检索结果-内蒙古大学图书馆

2024 International Conference on Computer Vision and Deep Learning, CVDL 2024

作者： Wei, Wei Zhang, Yi Du, Zexu Chu, Zongbo Zhang, Guoliang Grid Laboratory of Grid Advanced Computing and Applications State Grid Smart Grid Research Institute Co. Ltd. Beijing China

ISBN: (纸本)9798400718199

Safety control is one of the most important tasks for power grid operation. However, safety control and risk warning models with high accuracy always have multitudinous parameters and need scaled resources for calculation. So as to sort out this problem, we proposed a novel and adaptive model lightweight method based on knowledge distillation. In our method, the intermediate features of the teacher network is utilized to guide the learning of the student network and a regressor is introduced behind the selected middle layer of teacher network to realize dimension match. The prediction confidence of teacher network is mapped to the feature space for the regressor initialization. Besides, the prediction confidence is used to construct the distillation temperature to improve the adaptability of information transfer. Experiments on ImageNet, CIFAR 100 and Grid-specific datasets demonstrate the lightweight model obtained by our method possesses powerful and adaptive feature extraction and representation ability. © 2024 ACM.

关键词： Safety engineering

来源：评论

学校读者我要写书评

暂无评论

Enhancing Data Consistency in Distribution Networks Leveraging Graph Contrastive Representation Learning 6

Enhancing Data Consistency in Distribution Networks Leveragi...

引用

6th International Conference on Power and Energy Technology, ICPET 2024

作者： Liu, Siyan Shi, Danni Tang, Zhuang State Grid Laboratory of Grid Advanced Computing and Applications State Grid Smart Grid Research Institute Co.Ltd. Beijing China

ISBN: (纸本)9798350375855

The rapid integration of distributed renewable energy sources and the diversification of loads have precipitated profound changes in the configuration and operation of power distribution networks in the 21st century, underscoring the necessity for accurate and standardized power grid topology model data. This manuscript introduces a novel graph contrastive learning-based approach for the representation of node features in distribution networks, specifically targeting the alignment of identical nodes across different Common Information Model (CIM) documents. By combining self-supervised learning with graph neural networks, a feature extraction framework that integrates topological structural features and attribute features of the distribution network is proposed in the paper. Besides, the study designs a graph view generation strategy considering distribution network characteristics. The proposed method is validated by an 11-feeder distribution network dataset, and the results show the effectivity of the graph contrastive learning. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A CBL Estimation Method for Industrial Customers Based on K-Means and LSTM 11

A CBL Estimation Method for Industrial Customers Based on K-...

引用

11th China International Conference on Electricity Distribution, CICED 2024

作者： Zhang, Wensi Lei, Shuya Liu, Mohan Wei, Renjie Wang, Yishen State Grid Laboratory of Grid Advanced Computing and Applications China Electric Power Research Institute Co. Ltd. Beijing China

ISBN: (纸本)9798350368345

In pursuit of attaining the objective of 'carbon peaking and carbon neutrality', the structure of energy in China has undergone a great transformation. With the rapid development of renewable energy generation, the power grid has been confronted with new challenges. The reliability and flexibility of the grid have been affected. Demand response (DR), as an effective management method, has become a hotspot. Customer baseline load (CBL) serves as the standard for calculating economic compensation in DR. Hence, accurate estimation of CBL is a crucial aspect in ensuring the effective implementation of DR strategies. This paper proposes a CBL estimation method for industrial customers based on the K-means and long short-term memory (LSTM). First, the K-means method is used to cluster the daily load curves of various industries to yield the typical patterns of electricity consumption behaviors. Second, the LSTM neural network is established to estimate CBL. It is trained with historical load data of different industries and clustering results. Finally, the effectiveness of the novel method is validated via a real data set. © 2024 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

research on Key Technologies of Intelligent Detection of Substation Equipment Based on Cloud Edge Collaboration 24

Research on Key Technologies of Intelligent Detection of Sub...

引用

2024 International Conference on Computer Vision and Deep Learning, CVDL 2024

作者： Chang, Ke Zheng, Min Wu, Chunpeng State Grid Laboratory of Grid Advanced Computing and Applications State Grid Smart Grid Research Institute Co. Ltd Beijing China

ISBN: (纸本)9798400718199

The traditional manual inspection has some problems, such as strong subjectivity, time consuming and poor efficiency. With the remarkable progress of artificial intelligence, advanced communication and other technologies, and the wide application of deep learning models, the intensity of daily work has been reduced, but at the same time, new problems have been created. At present, the memory and computing power of the terminal hardware are limited, and when the model is deployed to the terminal side, the processing efficiency of the model is slow and the accuracy is low due to the limited hardware conditions. Aiming at the limited memory and computing power of the edge hardware, the lightweight network structure and model compression strategy are studied, and a channel structured pruning algorithm is proposed. A lightweight deep learning network model based on structured pruning of VGG-16 model was constructed. Experiments show that the optimized pruning model can obtain the same detection performance as the original model, and can reduce the size of the model by about 55.3% and increase the speed of 33FPS when the average accuracy mAP is lost by 8.1%, which provides effective support for the intelligent detection of substation equipment. © 2024 ACM.

关键词： computing power

来源：评论

学校读者我要写书评

暂无评论

ContinUNet: fast deep radio image segmentation in the Square Kilometre Array era with U-Net

引用

RAS Techniques and Instruments 2024年第1期3卷 315-332页

作者： Stewart, Hattie Birkinshaw, Mark Yeung, Siu-Lun Maddox, Natasha Maughan, Ben Thiyagalingam, Jeyan School of Physics University of Bristol HH Wills Physics Laboratory SciML Scientific Computing Department Research Complex at Harwell Rutherford Appleton Laboratory UKRI Centre for Doctoral Training in Artificial Intelligence Machine Learning & Advanced Computing Department of Physics Vivian Tower Swansea University

We present a new machine learning (ML)-driven source-finding tool for next-generation radio surveys that performs fast source extraction on a range of source morphologies at large dynamic ranges with minimal parameter tuning and post-processing. The construction of the Square Kilometre Array (SKA) radio telescope will revolutionize the field of radio astronomy. However, accurate and automated source-finding techniques are required to reach SKA science goals. We have developed a novel source-finding method, ContinUNet, powered by an ML segmentation algorithm, U-Net, that has proven highly effective and efficient when tested on SKA precursor data sets. Our model was trained and tested on simulated radio continuum data from SKA Science Data Challenge 1 and proved comparable with the state-of-the-art source-finding methods, PyBDSF and ProFound. ContinUNet was then tested on the MeerKAT International GHz Tiered Extragalactic Exploration Early Science data without retraining and was able to extract point-like and extended sources with equal ease;processing a 1.6 deg² field in <13 s on a supercomputer and ≈2 min on a personal laptop. We were able to associate components of extended sources without manual intervention with the powerful inference capabilities learnt within the network, making ContinUNet a promising tool for enabling science in the upcoming SKA era. © 2024 The Author(s).

关键词： Data Methods Evolution Galaxies Galaxies Machine Learning Radio Continuum Software

来源：评论

学校读者我要写书评

暂无评论

Intelligent indoor metasurface robotics

引用

National Science Review 2023年第8期10卷 20-32页

作者： Hanting Zhao Shengguo Hu Hongrui Zhang Zhuo Wang Hao Dong Philipp del Hougne Tie Jun Cui Lianlin Li State Key Laboratory of Advanced Optical Communication Systems and Networks School of Electronics Peking University Center on Frontiers of Computing Studies School of Computer Science Peking University The French National Centre for Scientific Research(CNRS) University of Rennes State Key Laboratory of Millimeter Waves Southeast University

Intelligent indoor robotics is expected to rapidly gain importance in crucial areas of our modern society such as at-home health care and factories. Yet, existing mobile robots are limited in their ability to perceive and respond to dynamically evolving complex indoor environments because of their inherently limited sensing and computing resources that are, moreover, traded off against their cruise time and payload. To address these formidable challenges, here we propose intelligent indoor metasurface robotics(I2MR),where all sensing and computing are relegated to a centralized robotic brain endowed with microwave perception; and I2MR's limbs(motorized vehicles, airborne drones, etc.) merely execute the wirelessly received instructions from the brain. The key aspect of our concept is the centralized use of a computation-enabled programmable metasurface that can flexibly mold microwave propagation in the indoor wireless environment, including a sensing and localization modality based on configurational diversity and a communication modality to establish a preferential high-capacity wireless link between the I2MR's brain and limbs. The metasurface-enhanced microwave perception is capable of realizing low-latency and high-resolution three-dimensional imaging of humans, even around corners and behind thick concrete walls, which is the basis for action decisions of the I2MR's brain. I2MR is thus endowed with real-time and full-context awareness of its operating indoor environment. We implement, experimentally, a proof-of-principle demonstration at ～2.4 GHz, in which I2MR provides health-care assistance to a human inhabitant. The presented strategy opens a new avenue for the conception of smart and wirelessly networked indoor robotics.

关键词： intelligent metasurfaces intelligent robotics edge devices IoT 6G

来源：评论

学校读者我要写书评

暂无评论

Masked Image Modeling Knowledge Distillation Based on Mutual Learning 24

Masked Image Modeling Knowledge Distillation Based on Mutual...

引用

2024 International Conference on Computer Vision and Deep Learning, CVDL 2024

作者： Zhang, Yi Du, Zexu Wei, Wei Chu, Zongbo State Grid Laboratory of Grid Advanced Computing and Applications State Grid Smart Grid Research Institute Co. Ltd. Beijing China

ISBN: (纸本)9798400718199

With the rapid improvement of the performance of large models, pre-training a large-scale model and fine-tuning on downstream has become a trend. However, previous knowledge distillation methods mainly concentrates on a specific downstream task. This paper proposes a novel distillation framework Mutual Distillation for Masked Autoencoder Model(MMIM) which transfers knowledge from large-scale pre-trained visual models to provide smaller backbones for downstream tasks, thereby saving time. In our model MMIM, to maintain the advantages of mutual learning, two different student models learn collaboratively to boost both feature representation capabilities simultaneously. At the same time, in order to utilize powerful feature representation capability of pre-trained teacher model, we use the intermediate features of the teacher model to supervise the learning of the student models. Experiments on ImageNet demonstrates the compressed model obtained by our framework possesses powerful feature representation comparable to teacher model. In addition, MMIM also outperforms state-of-the-art knowledge distillation methods which directly distills fine-tuned models. © 2024 ACM.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Vertex Centric Weighted Edge Index to Accelerate Weighted Graph Traversal 4

Vertex Centric Weighted Edge Index to Accelerate Weighted Gr...

引用

4th International Conference on Consumer Electronics and Computer Engineering, ICCECE 2024

作者： Lin, Jianchao Ren, Chunhui Liu, Siyan State Grid Smart Grid Research Institute Co. Ltd. State Grid Laboratory of Grid Advanced Computing and Applications Beijing China

ISBN: (纸本)9798350384062

Indexing is an important technique to optimize graph database performance. However, indexing in existing graph databases focuses mainly on property-value based query, and has not drawn much attention on graph traversal. We propose a new indexing technique, which maintains sorted indexing of vertex and weight values of the vertex's adjacent edges. With this proposed vertex centric adjacent weighted edge index, graph traversal algorithms based on weight values like Dijkstra's algorithm and Prim's algorithm have significant performance gains. © 2024 IEEE.

关键词： Graph Databases

来源：评论

学校读者我要写书评

暂无评论

Blockchain-Enabled Secure, Fair and Scalable Data Sharing in Zero-Trust Edge-End Environment

引用

IEEE Journal on Selected Areas in Communications 2025年第6期43卷 2056-2069页

作者： Xu, Xiaolong Meng, Ke Xiang, Haolong Cui, Guangming Xia, Xiaoyu Dou, Wanchun Nanjing210044 China Nanjing University of Information Science and Technology School of Software and the Jiangsu Province Engineering Research Center of Advanced Computing and Intelligent Services Nanjing210044 China RMIT University School of Computing Technologies MelbourneVIC3000 Australia Nanjing University State Key Laboratory for Novel Software Technology Nanjing210023 China

In edge computing, the Zero-Trust Security Model (ZTSM), as a key enabling technology for next-generation networks, plays a crucial role in providing authentication for addressing data sharing concerns, such as frequent data breaches, data misuse, and cyberattacks. However, due to the complexity and diversity of edge environments, ZTSM struggles to meet the security requirements of data sharing frameworks solely through enhanced authentication. Consequently, such frameworks with ZTSM still face challenges in ensuring data integrity, evaluating various node behaviors, and coping with the increasing complexity of node attributes. To address these issues, we propose a blockchain-enabled secure, fair and scalable data sharing framework in a zero-trust edge-end environment in this paper. Specifically, we first propose a Merkle forest-based data storage model for the classified storage of loosely coupled data, consequently enhancing the scalability of the model. Then, we design a node behavior-based reputation assessment mechanism to ensure fairness during data sharing. Moreover, a data sharing protocol supervised by smart contract is proposed, working with the aforementioned storage and assessment schemes, to ensure the security of data sharing. Finally, comprehensive security analysis validates the security, fairness and scalability of the proposed framework. Extensive experimental results show that, as transaction volume grows, the time cost of data traversal in the storage model becomes progressively more efficient. Additionally, when the size of the smart contract is increased tenfold, the maximum time cost of the data sharing protocol rises by only 4.98 times. © 1983-2012 IEEE.

关键词： Data integrity

来源：评论

学校读者我要写书评

暂无评论

Application of Self-Distillation Framework in Transmission Line Defect Recognition 7

Application of Self-Distillation Framework in Transmission L...

引用

7th IEEE advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2024

作者： Yang, He Shi, Cuncun Lin, Long State Grid Laboratory of Grid Advanced Computing and Applications State Grid Smart Grid Research Institute Co. Ltd. Beijing102209 China

ISBN: (纸本)9798350339161

Drones have become very common in inspections of transmission lines across the country, but the images and videos captured by drones still require manual processing, which is time-consuming and labor-intensive, not intelligent enough to meet real-time requirements, and the response is relatively passive;currently in the power sector artificial intelligence has been used to process drone footage, but it has not yet reached the level of human experts. In recent years, methods to improve model performance have focused on conceiving new network structures and improving the quantity and quality of sample sets. Both methods are costly. This paper proposes a training framework for the model based on self-distillation. Without changing the model structure or changing the data set, you can learn the knowledge of your own network to improve model performance and use smaller models under the same performance requirements. Based on this framework, we implemented a self-distillation system on ResNet and verified the framework by comparing original training and self-distillation training. Of course, this framework is also applicable to other networks. Experiments show that the self-distillation method can significantly improve the accuracy of recognition. © 2024 IEEE.

关键词： Electric lines

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：