检索结果-内蒙古大学图书馆

4th IEEE international conference on Power, Intelligent computing and Systems, ICPICS 2022

作者： Zhou, Feng Tang, Weijie Lu, Songheng Hu, Heng Jiang, Peng Hao, Ting College of Electrical Engineering and Automation Shandong University of Science and Technology Qingdao China Shandong Maternal and Child Health Hospital Jinan China Power Metering Center State Grid Heilongjiang Electric Power Company Ltd. Harbin China Electric Power Research Institute State Grid Heilongjiang Electric Power Company Ltd. Harbin China College of Economics and Management Shandong University of Science and Technology Qingdao China

ISBN: (数字)9781665467735

ISBN: (纸本)9781665467735

In view of the disadvantages of traditional global maximum power point tracking (GMPPT) algorithms such as particle swarm optimization and cuckoo search algorithm that the structure is complex, the tracking path is uncertain and the requirement on controller hardware is high, this paper adapts an improved cuckoo search algorithm to track the GMPP of series-parallel PV array. Compared with the traditional algorithm by analyzing the tracking process of maximum power point under uniform illumination and partial shading condition, the algorithm makes it faster and more accurate to track the GMPP of the series-parallel photovoltaic array connected to the DC distribution network under both conditions and finally provides the technical foundation for the safe and efficient connection of distributed photovoltaic power to the distribution network. © 2022 IEEE.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

Optimizing the parallelism of Communication and Computation in distributed Training Platform 23rd

Optimizing the Parallelism of Communication and Computation ...

引用

23rd international conference on Algorithms and Architectures for parallel Processing (ICA3PP)

作者： Hou, Xiang Yuan, Yuan Ma, Sheng Xu, Rui Wang, Bo Li, Tiejun Jiang, Wei Wu, Lizhou Zhang, Jianmin Natl Univ Def Technol Changsha Peoples R China

ISBN: (纸本)9789819708338;9789819708345

With the development of deep learning, DNN models have become more complex. Large-scale model parameters enhance the level of AI by improving the accuracy of DNN models. However, they also present more severe challenges to the hardware training platform for training a large model needs a lot of computing and memory resources, which can easily exceed the capacity of an accelerator. In addition, with the increasing demand for the accuracy of DNN models in academia and industry, the number of training iterations is also skyrocketing. In these backgrounds, more accelerators are integrated on a hierarchical platform to conduct distributed training. In distributed training platforms, the computation of the DNN model and the communication of the intermediate parameters are handled by different hardware modules, so their degree of parallelism profoundly affects the training speed. In this work, based on the widely used hierarchical Torus-Ring training platform and the Ring All-Reduce collective communication algorithm, we improve the speed of distributed training by optimizing the parallelism of communication and computation. Specifically, based on the analysis of the distributed training process, we schedule the computation and communication so that they execute simultaneously as much as possible. Finally, for data parallelism and model parallelism, we reduce the communication exposure time and the computation exposure time, respectively. Compared with the previous work, the training speed (including 5 training iterations) of the Resnet50 model and the Transformer model is increased by 23.77%-25.64% and 11.66%-12.83%.

关键词： distributed training simulator Collective communication operation Communication exposure time Computation exposure time

来源：评论

学校读者我要写书评

暂无评论

Multi-time scale electric vehicle charging load forecasting considering constant current charging and parallel computing

引用

ENERGY REPORTS 2022年 8卷 722-732页

作者： Bian, Haihong Guo, Zhengyang Zhou, Chengang Peng, Shan Nanjing Inst Technol 1 Hongjing AveJiangning Sci Pk Nanjing 211167 Peoples R China

At present, the concept of ecological civilization has been widely recognized by the whole world, and a series of policies that guarantees the evolution of the electric vehicle (EV) industry has been implemented by multiple countries. Therefore, it is significant to predict the charging load of EVs to solve the challenges of power system planning and operation. The Monte Carlo (MC) method is preferred by many scholars in EV charging load prediction because it is very suitable for describing random characteristics with a good prediction effect. To obtain more reliable and efficient prediction results, this paper analyzes the application of parallel computing technology in MC simulation. Firstly, EVs in the region are classified according to their battery capacity. Based on the voltage change curve of lithium-ion power batteries in the process of constant current charging under different capacities, the charging power, charging time, and state of charge (SOC) of EVs are investigated. Secondly, the behavior characteristics of users and the driving parameter characteristics of EVs are studied respectively, and the probability distribution model of multi-source information is established. Thirdly, parallel computing technology in the computer field is introduced, and an improved MC method is proposed based on the multi-core CPU architecture. After fully considering the complex constant current charging process after fitting, the charging load of EVs in a region of East China is simulated. Finally, the time cost and the load forecasting results of serial and parallel methods are compared and analyzed, and the progressiveness and effectiveness of the parallel method are verified. Results show that the charging load has four peaks in a day, taxis are the main source of the peak load of the power grid, the charging load of buses fluctuates the most, and private cars are the main backup capacity to participate in V2G dispatching in the future. In addition, under the experimen

关键词： Load forecasting Constant current charging process Electric vehicle Monte Carlo method parallel computing

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Resampling Technique with Adaptive Intervals Used in the parallel/distributed Particle Filters 7

A Hybrid Resampling Technique with Adaptive Intervals Used i...

引用

7th international conference on Computer Science and Artificial Intelligence, CSAI 2023

作者： Zhang, Xudong Gu, Feng Zhong, Wei Ai, Chunyu Department of Mathematics and Computer Science University of South Carolina Upstate SpartanburgSC 29303 United States Department of Computer Science The College of Staten Island Staten IslandNY10314 United States

ISBN: (纸本)9798400708688

parallel/distributed particle filters estimate the states of dynamic systems by using Bayesian interference and stochastic sampling techniques with multiple processing units (PUs). The sampling procedure and the resampling procedure alternatively execute to estimate the states in particle filters. There are two basic types of resampling techniques used in parallel/distributed particle filters. They are centralized resampling and decentralized resampling. The high communication between PUs in centralized resampling lowers the speedup factor in parallel computing but improves the estimation accuracy. The decentralized resampling can avoid the communication and improve the performance. Some types of hybrid resampling techniques mainly execute the decentralized resampling and only invoke the centralized resampling with constant intervals to achieve ideal performance without losing the estimation accuracy. However, the constant intervals cannot guarantee that the centralized resamplings are invoked timely. In this study, we proposed a hybrid resampling technique with adaptive intervals between centralized resamplings to overcome that issue. The experimental results indicate that the proposed hybrid resampling technique is able to improve the performance and the estimation accuracy. © 2023 ACM.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

parallel ATTRIBUTE COMPUTATION FOR distributed COMPONENT FORESTS 29

PARALLEL ATTRIBUTE COMPUTATION FOR DISTRIBUTED COMPONENT FOR...

引用

IEEE international conference on Image Processing (ICIP)

作者： Gazagnes, Simon Wilkinson, Michael H. F. Univ Texas Austin Dept Astron Austin TX 78712 USA Univ Groningen Bernouilli Inst Math & Artificial Intelligence Groningen Netherlands

ISBN: (数字)9781665496209

ISBN: (纸本)9781665496209

Component trees are powerful image processing tools to analyze the connected components of an image. One attractive strategy consists in building the nested relations at first and then deriving the components' attributes afterward, such that the user can switch between different attribute functions without having to re-compute the entire tree. Only sequential algorithms allow such an approach, while no parallel algorithm is available. In this paper, we extend a recent method using distributed memory techniques to enable posterior attribute computation in a parallel or distributed manner. This novel approach significantly reduces the computational time needed for combining several attribute functions interactively in Giga and Tera-Scale data sets.

关键词： Mathematical morphology Connected filters Component trees Image representation parallel computing

来源：评论

学校读者我要写书评

暂无评论

SPM-GCN: An adaptive reordering algorithm for sparse LU factorization via GCN 29

SPM-GCN: An adaptive reordering algorithm for sparse LU fact...

引用

29th IEEE international conference on parallel and distributed Systems, ICPADS 2023

作者： Tian, Min Liu, Huazeng Wang, Junjie Liu, Qi Du, Wei Zhang, Zanjun China Shandong Fundamental Research Center for Computer Science Jinan Key Laboratory of High Performance Computing Shandong Provincial Key Laboratory of Computer Networks China Xidian University Shaanxi Key Laboratory of Large Scale Electromagnetic Computing Xi'an710071 China

ISBN: (纸本)9798350330717

Sparse LU factorization is a critical kernel in scientific computing and engineering applications. A better nonzero pattern of sparse matrixes can accelerate LU factorization by reordering. Traditionally it's difficult to predict which non-zero pattern is optimal for a sparse matrix. In this paper, we proposed a graph convolutional neural network (GCN) for adaptively selecting the optimal reordering algorithm from five candidate reordering algorithms in the Matrix preprocessing step, referred to as SPM-GCN. After using our SPM-GCN, the average numerical factorization time outperformed the other five algorithms, and the average numerical factorization time was reduced by 17.3% compared with the default method of swSuperLU. © 2023 IEEE.

关键词： Graph Convolutional Network reordering algorithm sparse linear system solver Sparse LU factorization

来源：评论

学校读者我要写书评

暂无评论

Load Balancing of the parallel Execution of Two Dimensional Partitioned Cellular Automata 30

Load Balancing of the Parallel Execution of Two Dimensional ...

引用

30th Euromicro international conference on parallel, distributed and Network-Based Processing (PDP)

作者： Giordano, Andrea Amelia, Francesca Gigliotti, Salvatore Rongo, Rocco Spataro, William ICAR CNR Arcavacata Di Rende Italy Univ Calabria Dept Math & Comp Sci Calabria Italy Univ Calabria Dept Environm Engn Calabria Italy

ISBN: (纸本)9781665469586

Load Balancing is generally referred as the technique to properly partition computation among processing elements in order to achieve optimal resource usage and thus reduce computation time. In this paper, we present a dynamic load balancing application in the context of the parallel execution of Cellular Automata where the domain space is partitioned in two dimensional regions that are assigned to different processing elements. Starting from general closed-form expressions that allow to compute the optimal workload assignment in a dynamic fashion when partitioning takes place along only one dimension, we extend the procedure to allow partitioning and balancing along both dimensions. As confirmed by the experimental results, two dimensional partitioning itself enables to speedup the execution, and further improvements are obtained when the load balancing occurs along both dimensions.

关键词： Load Balancing parallel computing Cellular Automata

来源：评论

学校读者我要写书评

暂无评论

international conference on Electronics, Electrical and Information Engineering, ICEEIE 2024

International Conference on Electronics, Electrical and Info...

引用

2024 international conference on Electronics, Electrical and Information Engineering, ICEEIE 2024

ISBN: (纸本)9781510686793

The proceedings contain 118 papers. The topics discussed include: research on improving sagging control strategy based on distributed inverter;thermal electric coupling analysis of direct heating of oil shale by power supply;research on control strategy of grid-connected inverter for compressed air energy storage system;research on high performance analysis technology of distribution network digital twin based on parallel acceleration;research on the feedback braking optimization system of electric vehicle that makes full use of ground adhesion;lightweight method for cognitive large model of electric power equipment operation and inspection based on knowledge distillation;research on intelligent control algorithms and applications for mechatronic systems of robots;and research on the application of computer vision technology in automatic detection of substation equipment.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ultra-Short-Term Load Forecasting Based on EMD Decomposition and CNN-BiGRU Heterogeneous computing Model 4

Ultra-Short-Term Load Forecasting Based on EMD Decomposition...

引用

4th IEEE international conference on Power, Electronics and Computer Applications, ICPECA 2024

作者： Dong, Yun Meng, Qi Fu, Jiacheng Ai, Xuhua Chen, Zhaoli Yin, Yuan Liu, Kaijie Zhang, Xixiang Guangxi Power Grid Co. Ltd Nanning China

ISBN: (纸本)9798350359589

Accurate and effective load forecasting holds utmost importance in ensuring the secure and stable operation of power grids. To bolster the precision and efficiency of load forecasting models, this research integrates the Empirical Mode Decomposition (EMD) algorithm to decompose load data into distinct frequency components. Additionally, a novel CNN-BiGRU deep learning fusion model is introduced to capture intricate sequence dependencies within the data. This involves directing different component features to separate models for individualized training. In a bid to optimize efficiency, the study leverages a parallel acceleration platform for streamlined model building and training processes. Subsequently, prediction outcomes from diverse models are amalgamated and synthesized, culminating in the attainment of final load prediction results. To validate the model's efficacy, the approach is assessed through a case study employing load data from a specific station area. Comparative analysis with models like BiGRU and CNN-GRU is conducted. The findings underscore the high forecasting accuracy achieved by the model developed in this study, affirming its capacity to facilitate high-quality ultra-short-term load forecasting. © 2024 IEEE.

关键词： EMD decomposition heterogeneous computing Load forecasting neural network

来源：评论

学校读者我要写书评

暂无评论

A Flexible Allocation of distributed Energy Sources Coordinated with Network Switching and Capacitor Placement Under System Voltage-Dependent Loading

A Flexible Allocation of Distributed Energy Sources Coordina...

引用

2023 IEEE international conference on Power Electronics, Smart grid, and Renewable Energy, PESGRE 2023

作者： Mahdavi, Meisam Awaafo, Augustine Jurado, Francisco University of Jaén Department of Electrical Engineering Jaén Linares Spain

ISBN: (纸本)9798350310573

Sufficient reduction in energy loss of distribution networks is achievable through simultaneous network switching (NS), capacitor banks (CBs) placement, and the control of distributed energy sources (DESs), where the role of load power is so important in the process. Therefore, accurately modeling the load is essential in relation to the analyses because the load behavior of constantly varying loads with voltage magnitudes affects network losses and, subsequently, the optimal controls for NS, CBs, and DESs. Regarding the fact that load presentation in terms of constant current, impedance, and power components is practically important, this paper analyzes the relationship of load components and load types in order to predict more reliable coordinated NS, CB, and DES controls for losses reduction. The analysis is implemented on the 16-bus distribution network using the mathematical formulation of load components in terms of consumer type. The obtained results proved that the proposed framework is optimal for obtaining an accurate solution in a relatively short time as compared to exact models of NS, CB, and DES controls. © 2023 IEEE.

关键词： Capacitor placement consumption components distributed energy sources flexibility network switching

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：