检索结果-内蒙古大学图书馆

Semi-asynchronous personalized federated learning for short-term photovoltaic power forecasting

Digital communications and Networks 2023年第5期9卷 1221-1229页

作者： Weishan Zhang Xiao Chen Ke He Leiming Chen Liang Xu Xiao Wang Su Yang College of Computer Science and Technology China University of Petroleum(East China)China Sichuan Energy Internet Research Institute Tsinghua UniversityChina School of Computer and Communication Engineering Beijing University of Science and TechnologyChina State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijingChina School of Computer Science Fudan UniversityChina

Accurate forecasting for photovoltaic power generation is one of the key enablers for the integration of solar photovoltaic systems into power *** deep-learning-based methods can perform well if there are sufficient training data and enough computational ***,there are challenges in building models through centralized shared data due to data privacy concerns and industry *** learning is a new distributed machine learning approach which enables training models across edge devices while data reside *** this paper,we propose an efficient semi-asynchronous federated learning framework for short-term solar power forecasting and evaluate the framework performance using a CNN-LSTM *** design a personalization technique and a semi-asynchronous aggregation strategy to improve the efficiency of the proposed federated forecasting *** evaluations using a real-world dataset demonstrate that the federated models can achieve significantly higher forecasting performance than fully local models while protecting data privacy,and the proposed semi-asynchronous aggregation and the personalization technique can make the forecasting framework more robust in real-world scenarios.

关键词： Photovoltaic power forecasting Federated learning Edge computing CNN-LSTM

来源：评论

学校读者我要写书评

暂无评论

Strip Weld Inspection Based on Improved VGG16 8

Strip Weld Inspection Based on Improved VGG16

引用

8th International Conference on Signal and Image Processing, ICSIP 2023

作者： Fu, Sisi Chen, Yuan Xian, Dexin Kong, Shiyuan Wang, Hao Wang, Jinsong Changchun University of Science and Technology School of Optoelectronic Engineering Guangdong Zhongshan China Harbin University of Science and Technology School of Measurement and Control Technology and Communication Engineering Heilongjiang Harbin China Zhongshan Institute School of Optoelectronic Engineering Changchun University of Science and Technology Guangdong Zhongshan China Changchun University of Science and Technology Zhongshan Institute Changchun University of Science and Technology School of Optoelectronic Engineering Guangdong Zhongshan China

ISBN: (纸本)9798350397932

As one of the indispensable materials in modern industry, high precision and high efficiency defect detection of strip weld has become an important part of production and quality control. In response to the shortcomings of the existing strip weld detection technology with low detection accuracy, this paper proposes a deep learning based strip weld defect detection and classification technique. First, the CBAM attention mechanism is introduced into the network structure to improve the attention to important features;second, the traditional convolution layer is replaced by using the dilated convolution to reduce the computation while increasing the perceptual field;finally, the original ReLU activation function is replaced by the PReLU activation function to effectively overcome the problem of neuron death in the Relu activation function. The improved model finally achieves a recognition rate of 98.1% in the test set, which is 22.1% higher than that of the unimproved model, greatly improving the accuracy and robustness of the algorithm. © 2023 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Collision-Free Formation control of Multi-agent System with Low Oscillation 7th

Collision-Free Formation Control of Multi-agent System with ...

引用

7th Chinese Conference on Swarm Intelligence and Cooperative control, CCSICC 2023

作者： Li, Jingyi Han, Haoran Lv, Maolong Sun, Chenyang Cheng, Jian Glassgow School University of Electronic Science and Technology of China Chengdu611731 China School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu611731 China Air Traffic Control and Navigation College Air Force Engineering University Xian710051 China

ISBN: (纸本)9789819733279

In recent years, multi-agent systems are becoming more complex, scalable, adaptive, and integrating machine learning for enhanced capabilities. A popular approach for addressing this challenge involves the utilisation of artificial potential functions (APF). In this article, we address the problem of obstacle avoidance and achieving a predefined formation, particularly promoting smooth trajectory generation in multi-agent formation control. We employ the APF method to enhance the smoothness of the agents’ movement paths during these tasks. Collision damping terms are generated by differentiating the gradient of APF with respect to time. Incorporating generated damping terms in the APF control, we suppress irregular fluctuations during agents’ motion procedures. To ensure stability in the multi-agent system, we design an energy-based Lyapunov function and utilise it for proof. Simulation examples are presented to validate the theoretical results and demonstrate the advantages of the proposed control algorithm. Through these simulations, it becomes evident that the collision-free formation control algorithm enhances the smoothness of the agents’ trajectories. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Robust Federated Learning Algorithm Based on Chaotic Encryption and Dual-Server Detection

引用

IEEE Internet of Things Journal 2025年

作者： Kang, Shouqiang Wu, Yuxuan Wang, Yujing Wang, Qingyan Liang, Xintao Harbin University of Science and Technology School of Measurement and Control Technology and Communication Engineering Harbin150080 China

Federated learning allows users to collaboratively train models, but studies have shown that local model parameters may leak users’ privacy. Furthermore, the global model in federated learning could be compromised by Byzantine attacks. To address these issues, a robust federated learning algorithm based on chaotic encryption and dual-server detection (CDRFL) is proposed. First, CDRFL introduces a lightweight chaotic encryption method, where ciphertexts have a unique property of mutual cancellation, allowing the plaintext global model parameters to be derived without decryption. Moreover, CDRFL allows users to encrypt with different keys, further reducing the risk of privacy leakage. Next, a dual-server detection scheme is designed to defend against Byzantine attacks while preserving user privacy. Lastly, a reputation score mechanism with a forgetting function is proposed to fairly distinguish Byzantine participants. Experimental results demonstrate that, compared with popular robust aggregation schemes, CDRFL improves testing accuracy by 3.26% to 9.06%. When compared to existing chaotic encryption federated learning schemes, CDRFL enhances the ability to defend against Byzantine attacks and reduces time and communication overhead by 73.63% and 95.40%, respectively. © 2014 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Enhanced Responsivity of Solar Blind Ultraviolet Photodetector by PEALD Deposited Zn-Doped Ga2O3Thin Films

引用

IEEE Transactions on Electron Devices 2024年第1期71卷 664-669页

作者： Fan, Hui-Chen Wang, Chen Ruan, Yu-Jiao Shen, Kun-Ching Wu, Wan-Yu Wuu, Dong-Sing Lai, Feng-Min Lien, Shui-Yang Zhu, Wen-Zhang Xiamen Institute of Measurement and Testing National Measurement and Testing Center for Flat Panel Display Industry Xiamen361024 China Fuzhou University College of Physics and Information Engineering Fuzhou350108 China National United University Department of Materials Science and Engineering Miaoli36063 Taiwan National Chi Nan University Department of Applied Materials and Optoelectronic Engineering Nantou54561 Taiwan Xiamen University of Technology Xiamen Key Laboratory of Development and Application for Advanced Semiconductor Coating Technology School of Opto-Electronic and Communication Engineering Xiamen361026 China Da-Yeh University Department of Biomedical Engineering Changhua51591 Taiwan

In this article, the tunable Zn-doped Ga2O3 films were obtained by introducing ZnO intercalation during the film deposition using plasma-enhanced atomic layer deposition (PEALD). The effect of the ZnO cycle ratio on the performance of solar-blind photodetectors (PDs) was first analyzed. Compared to the pure Ga2O3, an obviously enhanced performance of Zn-doped Ga2O3 films PDs was observed by the introduction of Zn dopants, especially photocurrent or responsivity, even if the light intensity is low. A high sensitivity of Zn-doped Ga2O3 PD has been achieved with superior photoelectric characteristics for an extremely low dark current of 2.22 × 10-13 A, an ultrahigh light ON/ OFF current ratio of 2.89 × 106, and a satisfactory responsivity of 104 mA/W when the ZnO cycle ratio is 5%. The above excellent results imply that the Zn-doped Ga2O3 PD can be greatly beneficial to the application in deep-ultraviolet (DUV) detection. © 1963-2012 IEEE.

关键词： II-VI semiconductors

来源：评论

学校读者我要写书评

暂无评论

Wrapper-based feature selection for general data set: the quantum sand cat swarm optimization

Wrapper-based feature selection for general data set: the qu...

引用

Computer, Vision and Intelligent technology (ICCVIT), International Conference on

作者： Jin Yue Ying Guo Hongyuan Gao Dalian Institute of Measurement and Control Technology Dalian China College of Information and Communication Engineering Harbin Engineering University Harbin China

ISBN: (数字)9798331540043

ISBN: (纸本)9798331540050

The existing wrapper feature selection methods have problems such as the difficulty of balancing a relatively high classification accuracy rate with a low feature selection rate, and there are problems such as long time consumption. In response to these issues, this paper combines quantum computing theory and sand cat swarm optimization to design quantum sand cat swarm optimization and combined with K-nearest neighbor method for feature selection. The optimal solution of quantum sand cat swarm optimization is set as the selected optimal feature subset. Simulation experiments are executed on 6 UCI data sets to compare with 4 classical swarm intelligence algorithms. The simulation outcomes show that the wrapper feature selection on the ground of quantum sand cat swarm optimization achieves lower feature selection rate , faster time and higher classification accuracy than the comparison method. It is a high precision and practical wrapper feature selection method.

关键词： Quantum computing Accuracy High dimensional data Quantum mechanics Nearest neighbor methods Feature extraction Robustness Classification algorithms Particle swarm optimization Optimization

来源：评论

学校读者我要写书评

暂无评论

Sparse Recovery Space Time Adaptive Processing Based on Log-Determinant 15

Sparse Recovery Space Time Adaptive Processing Based on Log-...

引用

15th International Conference on Signal Processing Systems, ICSPS 2023

作者： Chang, Yuqing Du, Xiaolin Jin, Biao Tian, Tuanwei Li, Jianbo School of Computer and Control Engineering Yantai University Yantai China Ocean College Jiangsu University of Science and Technology Zhenjiang China School of Physics and Electronics Henan University Kaifeng China School of Communication and Information Engineering Chongqing University of Posts and Telecommunications Chongqing China

ISBN: (纸本)9781510675056

Airborne radars usually face non-uniform clutter environments, and it is difficult to obtain enough independent and identical distributed (i.i.d.) training samples, which degrades the clutter suppression performance of space time adaptive processing (STAP). To address the problem, a log-determinant sparse rocovery based STAP (LSR-STAP) method is proposed in this paper. Through the sparse recovery theory of low-rank matrices and the prior knowledge of clutter covariance matrices, the corresponding problem is modeled using the log-determinant (LogDet) approximation of the rank function, and the solution of the resulting nonconvex optimization problem is derived in the framework of the symmetric alternating direction method of multipliers (S-ADMM). The simulation results show the superiority of the proposed method over similar algorithms. © 2024 SPIE.

关键词： Space time adaptive processing

来源：评论

学校读者我要写书评

暂无评论

MULTI-ROBOT PATH PLANNING AND COOPERATIVE OBSTACLE AVOIDANCE BASED ON MONTE-CARLO TREE SEARCH 10

MULTI-ROBOT PATH PLANNING AND COOPERATIVE OBSTACLE AVOIDANCE...

引用

10th International Symposium on Test Automation and Instrumentation, ISTAI 2024

作者： Dai, Juan Jing, Yanzhang Su, Zhong Zhu, Cui Beijing Key Laboratory of High Dynamic Navigation Technology Beijing Information Science & Technology University Beijing China Key Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing China School of Automation Beijing Information Science & Technology University Beijing China School of Information and Communication Engineering Beijing Information Science & Technology University Beijing China

ISBN: (纸本)9781837242672

With the rapid development of artificial intelligence technologies, multi-robot cooperative processing tasks are becoming more and more common in various applications. However, traditional single-robot planning algorithms show obvious limitations in dealing with complex social interactions and dynamic environments in the real world. To address this challenge, this paper applies Monte Carlo Tree Search (MCTS) to multi-robot path planning and proposes a collaborative obstacle avoidance motion planning algorithm based on MCTS. Specifically, the algorithm prunes the Monte Carlo tree by excluding cases that do not meet the robot speed constraints and potential collisions, which significantly improves the computational speed and success rate of the algorithm, making it applicable to the multi-robot path conflict problem. In addition, we design a policy space that conforms to the robot's operation and a cost function that meets the requirements of robot path planning to plan an efficient cooperative path for each robot. Through a series of simulation experiments, we demonstrate the algorithm's path planning and obstacle avoidance performance in a variety of conflict scenarios. The results show that the method is highly cooperative and effective in complex environments. © The Institution of engineering & technology 2024.

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

FPGA-Oriented Target Detection Accelerator Design

FPGA-Oriented Target Detection Accelerator Design

引用

Frontiers technology of Information and Computer (ICFTIC), IEEE International Conference on

作者： Weizhe Huang School of Measurement and Control Technology and Communication Engineering Harbin University of Science and Technology Harbin China

Target detection is widely applied in fields such as face recognition, autonomous driving, and industrial automation. However, when deploying target detection models based on convolutional neural networks on resource-constrained embedded platforms, issues like high computational complexity, long execution times, and significant on-chip memory usage arise. To address these challenges, this paper presents a target detection accelerator design based on Field Programmable Gate Array (FPGA). Firstly, the computational workload of convolutional neural networks is reduced by merging convolutional layers with batch normalization layers. Secondly, to mitigate on-chip storage space concerns, on-chip data fixed-point quantization is employed to decrease on-chip memory consumption. Subsequently, to tackle the issue of long execution times, a convolution calculation module based on the Winograd algorithm is designed, which reduces computation latency. Finally, a dual-buffer pipeline architecture is devised to optimize memory bandwidth. Experimental results demonstrate that the proposed solution accelerates YOLOv4-tiny on the ZYNQ-7020, achieving a performance of 21.35 GOP/s. The energy efficiency is 32.56 times higher than CPU and 9.76 times higher than GPU. Additionally, the convolution calculation module in this paper reduces inference time by 15% compared to the sliding window convolution calculation method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Cross-Domain Few-Shot Learning With Spectral-Spatial Split-Attention For Hyperspectral Image Classification

Cross-Domain Few-Shot Learning With Spectral-Spatial Split-A...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Peng Luo Qingyan Wang Junping Zhang Shouqiang Kang School of Measurement-Control and Communication Engineering Harbin University of Science and Technology Harbin China School of Electronics and Information Engineering Harbin Institute of Technology Harbin China

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Hyperspectral image classification (HSIC) is a pivotal technology in hyperspectral remote sensing, playing a widespread role in remote sensing applications. However, the limited number of labeled samples has always made hyperspectral image classification difficult. In response to this issue, researchers have delved into cross-domain classification studies. Moreover, significant progress on cross-domain HSIC has been made in recent years. Nevertheless, existing methods exhibit shortcomings, including inadequate exploitation of spectral and spatial information and a slow training speed, rendering them unsuitable for downstream application tasks. To address these challenges, this paper introduces a model of cross-domain few-shot learning with spectral-spatial split attention(S 3 A-CFSL). Channel attention and split attention are presented to emphasize effective spectral and spatial information for HSIC adaptively. Additionally, the ResNet variant, called ResNeSt, is employed to expedite the training speed of the model. Experimental results demonstrate notable enhancements in the proposed method's classification accuracy and model training speed across two public datasets.

关键词： Training Adaptation models Accuracy Feature extraction Rendering (computer graphics) Data mining Few shot learning

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：