检索结果-内蒙古大学图书馆

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Yingnan Tong Jun Qiu Hongye Lin Yongbin Wang School of Control Sciences and Engineering Zhejiang University Hang Zhou China School of Information Science and Engineering NingBoTech University NingBo China School of Electrical Engineering and Automation Jiangxi University of Science and Technology Ganzhou China

Most factories use automatic dispensing machines for workpiece sealing, and the processing efficiency is very high. The traditional manual detection is inefficient and costly. Besides, the rate of false detection and missed detection of defect products remains high. In view of this problem, this paper proposes a glue line defect detection model based on Yolov5 and the self-built workpiece defect data set is used for the research. Firstly, some conventional convolution modules in Backbone network are replaced with the corresponding deformable convolution modules to improve the generalization ability of the model. Then in order to improve the feature extraction ability of the model, the coordinate attention module is embedded in the Backbone network. The results show that the mean average precision of the proposed algorithm can reach 95.62%, which is greatly higher than initial Yolov5 algorithm. And the detection rate reaches 28.67 FPS, which can meet the real time detection.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Photovoltaic Power Prediction of Meteorological Based on Data Drive and Mechanism Model

Photovoltaic Power Prediction of Meteorological Based on Dat...

引用

Chinese control and Decision Conference, CCDC

作者： Shanna Luo Kaixiang Peng Changbin Hu Xuecheng Li School of Automation and Electrical Engineering University of Science and Technology Beijing Beijing China School of Electrical and Control Engineering North China University of Technology Beijing China

The proportion of photovoltaics (PV) in the grid has increased significantly with the extensive use of renewable energy. However, natural factors such as weather have a great influence on the output power of PV. Affected by the randomness and volatility of PV output power, there are unpredictable power disturbances, which can easily cause a mismatch between PV output and load demand, resulting in bus voltage fluctuations, thereby affecting the safety and stability of the system. Traditional PV power prediction models usually directly uses the data characteristics such as horizontal solar radiation flux and temperature in the meteorological forecast, ignoring the effect of solar angle on power prediction. In order to further improve the accuracy of PV power prediction, a prediction method based on the meteorological data driven and mechanism integration is proposed. Firstly, the influence mechanism of solar angle on power prediction is analyzed using the solar radiation bevel conversion principle, and the solar angle solving model is established. Secondly, the meteorological forecast data is expanded on the basis of the solar angle mechanism, the point forecast and the trend forecast models are established, and the prediction results are fused with the error index. Finally, through the test of different prediction time standards for 1 day, 3 days, and 7 days of the data of multiple real light fields, the effectiveness of the method proposed in this article was verified.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust Dynamic Latent Variable Model for Dynamic Process Monitoring With Missing Data in Cyber-Physical Systems

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on Industrial Cyber-Physical Systems 2023年 1卷 227-235页

作者： Le Zhou Lixinqian Yu Beiping Hou Hongbo Zheng Zheng-Guang Wu School of Automation and Electrical Engineering Zhejiang University of Science & Technology Hangzhou China Faculty of Engineering and IT University of Melbourne Melbourne Australia National Laboratory of Industrial Control Technology Institute of Cyber-Systems and Control Zhejiang University Hangzhou China

Although remarkable studies on dynamic process modeling and monitoring in cyber-physical systems have been conducted, there are still several limitations to these approaches, rendering them unsuitable for practical applications. Firstly, the common dynamic models are built based on complete data, which means that their modeling performance usually deteriorates when there are missing measurements. Secondly, practical data often exhibits more complex characteristics, such as non-stationary and non-Gaussian noises. Therefore, a robust dynamic latent variable model is proposed to address the challenges of dynamic process modeling and monitoring in complex measurement environments with missing values. The proposed method effectively handles non-Gaussian noises and recursively estimates missing values, thus enhancing the monitoring performance within the dynamic latent variable modeling framework. Finally, the feasibility and superiority of the proposed method are evaluated through a numerical example and a real wastewater treatment case, demonstrating its effectiveness and practicality in real-world scenarios.

关键词： Cyber-physical systems Principal component analysis Process monitoring Data models Load modeling Numerical models Loading

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning-Based Cooperative Guidance Law with Multi-Constraint

Deep Reinforcement Learning-Based Cooperative Guidance Law w...

引用

2023 China automation Congress, CAC 2023

作者： Gao, Zhihao Fan, Junfang Dou, Denghui Ji, Yi Li, Yafeng Wu, Jinmin School of Automation Beijing Information Science and Technology University Beijing China Beijing Key Laboratory of High Dynamic Navigation Technology Beijing China School of Aerospace Engineering Beijing Institute of Technology Beijing China Beijing Key Laboratory of Uav Autonomous Control Beijing Institute of Technology Beijing China

ISBN: (纸本)9798350303759

To enhance the destructive capabilities of small air-to-surface missiles, it is necessary to address the challenges related to terminal attack angle constraints and multi-missile coordinated attacks, we propose a cooperative guidance law with multi-constraint based on deep reinforcement learning in this paper. Firstly, an intelligent control term is introduced into the biased proportional navigation law with angle constraints. Secondly, a reward function is designed to reduce terminal attack angle error and flight time error, ensuring that the missile successfully reaches the target within the expected flight time, while simultaneously converging the attack angle to the desired value. Finally, the deep deterministic policy gradient algorithm is employed for the purpose of training and generating intelligent guidance instructions. Simulation results substantiate the efficacy of the devised guidance method in achieving accurate target strikes, both for individual missiles and in scenarios involving multiple missiles. And exhibiting better stability and guidance accuracy compared to the traditional biased proportional navigation law. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Atmospheric Two-Way Time Transfer over Kilometers Free-Space Link with Dual Optical Frequency Comb 22

Atmospheric Two-Way Time Transfer over Kilometers Free-Space...

引用

22nd IEEE International Conference on Communication Technology, ICCT 2022

作者： Huang, Wantao Chen, Changfeng Hou, Dong College of Science Kaili University Kaili China Southwest China Research Institute of Electronic Equipment Science and Technology on Electronic Information Control Laboratory Chengdu China School of Automation Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9781665470674

In this paper, we report an outdoor free space optical two-way time transfer technique with dual optical frequency comb (OFC). Using the transfer method, we transmitted the pulse time signal on the kilometer-level free space link. In the time transfer experiment, the time drift and time deviation of 1 km and 20 km links are measured and simulated respectively. The experimental results show that the total root mean square (RMS) time drift of 1-km link is about 2.2 ps within 5000 seconds, the time deviation is 2 ps when the average time is 1 second, and the time deviation is 0.32 ps when the average time is 100 s. The emulation estimates that the RMS time drift of a 20 km link is about 232 ps within 5000 seconds, the time deviation is 115 ps when the average time is 1 second, and the time deviation is 36 ps when the average time above 100 s. © 2022 IEEE.

关键词： Natural frequencies

来源：评论

学校读者我要写书评

暂无评论

A Deep Reinforcement Learning-Based Guidance Law with Impact Angle Constraint

A Deep Reinforcement Learning-Based Guidance Law with Impact...

引用

2023 China automation Congress, CAC 2023

作者： Wang, Xu Fan, Junfang Dou, Denghui Ji, Yi Li, Yafeng Wu, Jinmin School of Automation Beijing Information Science and Technology University Beijing China Beijing Key Laboratory of High Dynamic Navigation Technology Beijing China School of Aerospace Engineering Beijing Institute of Technology Beijing China Beijing Key Laboratory of Uav Autonomous Control Beijing Institute of Technology Beijing China

ISBN: (纸本)9798350303759

Addressing the challenge of ineffective impact angle control when employing conventional biased proportional navigation guidance for precision-guided missiles against maneuvering targets, this study proposes a guidance law with impact angle constraint by integrating deep reinforcement learning algorithms with biased proportional navigation guidance. The algorithm enhances traditional proportional navigation guidance by introducing a target maneuver compensation bias term. This bias term is treated as an action in a learning framework and optimized using the Proximal Policy Optimization algorithm. Simulation experiments validate that the proposed algorithm, compared to conventional proportional navigation guidance, not only reduces miss distance and impact angle error but also shortens flight time. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Design and Optimization of Liquid State Machine for Handwritten Digit Recognition

Design and Optimization of Liquid State Machine for Handwrit...

引用

2022 Workshop on Electronics Communication engineering

作者： Hou, Aihu Liu, Jie Zhou, Kaibo Zhou, Wenli School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China School of Civil and Hydraulic Engineering Huazhong University of Science and Technology Wuhan430074 China Hubei Key Laboratory of Digital River Basin Science and Technology Huazhong University of Science and Technology Wuhan430074 China Key Laboratory of Image Processing and Intelligent Control Huazhong University of Science and Technology Ministry of Education Wuhan430074 China School of Optical and Electronic information Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9781510666634

The most commonly used implementation of handwritten digit recognition based on convolutional neural networks requires equipment with high computing power, which is not suitable for edge devices. Recently, spiking neural network (SNN) has received more attentions due to its low power consumption and real-time performance, but training SNN is very difficult. As a special SNN, liquid state machine (LSM) has the advantages of simple structure and easy training, so it is very suitable for handwritten digit recognition on edge devices. But it has no advantage in recognition accuracy. In order to improve the performance of LSM, its reservoir needs to be optimized. In this paper, an efficient local optimization strategy is proposed, improving the recognition accuracy of LSM by 11.8% with less the training time. In order to reduce the runtime, auto-encoder and feature screening are used to compress the input handwritten digit image. After feature compression, the input storage is reduced by 57%, and the runtime is reduced by 30%. This work provides an effective way to realize handwritten digit recognition on the edge devices. © 2023 SPIE.

关键词： Computing power

来源：评论

学校读者我要写书评

暂无评论

Discretizing Continuous Action Space With Unimodal Probability Distributions for On-Policy Reinforcement Learning

引用

IEEE Transactions on Neural Networks and Learning Systems 2024年第6期36卷 11285-11297页

作者： Yuanyang Zhu Zhi Wang Yuanheng Zhu Chunlin Chen Dongbin Zhao Department of Control Science and Intelligence Engineering School of Management and Engineering and the Laboratory of Data Intelligence and Interdisciplinary Innovation Nanjing University Nanjing China Department of Control Science and Intelligence Engineering School of Management and Engineering and the Research Center for Novel Technology of Intelligent Equipment Nanjing University Nanjing China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China

For on-policy reinforcement learning (RL), discretizing action space for continuous control can easily express multiple modes and is straightforward to optimize. However, without considering the inherent ordering between the discrete atomic actions, the explosion in the number of discrete actions can possess undesired properties and induce a higher variance for the policy gradient (PG) estimator. In this article, we introduce a straightforward architecture that addresses this issue by constraining the discrete policy to be unimodal using Poisson probability distributions. This unimodal architecture can better leverage the continuity in the underlying continuous action space using explicit unimodal probability distributions. We conduct extensive experiments to show that the discrete policy with the unimodal probability distribution provides significantly faster convergence and higher performance for on-policy RL algorithms in challenging control tasks, especially in highly complex tasks such as Humanoid. We provide theoretical analysis on the variance of the PG estimator, which suggests that our attentively designed unimodal discrete policy can retain a lower variance and yield a stable learning process.

关键词： Task analysis Probability distribution Aerospace electronics Reinforcement learning Gaussian distribution Technological innovation Learning systems

来源：评论

学校读者我要写书评

暂无评论

Formal control Synthesis via Safe Reinforcement Learning Under Real-Time Specifications

Formal Control Synthesis via Safe Reinforcement Learning Und...

引用

International Conference on control and automation (ICCA)

作者： Peng Lv Guangqing Luo Zhou He Xianwei Li Xiang Yin Department of Automation Key Laboratory of System Control and Information Processing Shanghai Jiao Tong University Shanghai China School of Electrical and Control Engineering Shaanxi University of Science and Technology Xi'an China

ISBN: (数字)9798350354409

ISBN: (纸本)9798350354416

In recent years, reinforcement learning techniques have gained widespread application in control synthesis. However, in the context of safety-critical systems, employing trial-and-error based reinforcement learning may be unacceptable due to the potential risks it poses during the learning process. Consequently, the development of safe reinforcement learning techniques has become imperative. This paper addresses the challenge of safe reinforcement learning for controller synthesis, particularly when safety specifications are intricately linked to the real-time behavior of the system. To articulate time-sensitive requirements, we leverage Metric Interval Temporal Logic (MITL). To ensure safety throughout the learning process, we introduce an additional reactive controller called a shield. Specifically, the shield functions to reject any behavior that violates the real-time specifications, thus mitigating potential risks. The efficacy of our proposed approach is demonstrated through simulation results, highlighting its ability to satisfy safety constraints in the dynamic environment of controller synthesis.

关键词： Measurement Heuristic algorithms Simulation Process control Reinforcement learning Real-time systems Safety

来源：评论

学校读者我要写书评

暂无评论

AGV Self-Adaptive Decoupling PID controller Based on BAS Algorithm

AGV Self-Adaptive Decoupling PID Controller Based on BAS Alg...

引用

2022 Chinese automation Congress, CAC 2022

作者： Zhang, Chengpeng Lv, Shichao Zhang, Qiang Cui, Peng Peng, Kai Jiang, Xiangyuan Shandong University of Science and Technology College of Electrical Engineering and Automation Qingdao China College of Electrical Engineering and Automation Shandong University of Science and Technology Qingdao Borsen Intelligent Technology Co. Ltd. Qingdao China School of Control Science and Engineer Shandong University Jinan China School of Power and Energy Northwestern Polytechnical University Xi'an China Shandong University Qingdao China

ISBN: (纸本)9781665465335

Automated Guided Vehicle (AGV), serving as an important tool for wide industrial applications, suffers many control problems for coupling impacts of rotation and shift moving. We present a decoupling controller to improve the kinematic performance of AGV in the paradigm of PID control. For adaptive tuning of PID, a heuristic optimization strategy, named Beetle Antennae Search (BAS), is employed to generate proper adjustment of parameter estimation. Finally, the effectiveness of the algorithm is proved by simulation and experimental results. © 2022 IEEE.

关键词： Automatic guided vehicles

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：