检索结果-内蒙古大学图书馆

parallel computing-oriented method for long-time duration problem of force identification

ENGINEERING WITH COMPUTERS 2022年第2期38卷 919-937页

作者： Pan, Chudong Deng, Xiongjie Huang, Zhenjie Guangzhou Univ Sch Civil Engn Guangzhou Peoples R China

Rapid identification of dynamic forces is an important research subject. To reduce the computing time, a parallel computing-oriented method is proposed in this study for dealing with a long-time duration problem of force identification. The proposed method is implemented via three continues steps, i.e., partition of parallel computing tasks, solution of parallel computing tasks and fusion of the identified results. In the first step, moving time window is applied for splitting an original problem into several sub-problems in time domain. Then the next step focuses on solving the sub-problems which can be executed in parallel. Herein, influences of unknown initial conditions are considered. Sparse regularization such as weightedl(1)-norm regularization method is introduced for ensuring that the identified result is sparse and stable. In the last step, the identified results calculated from all the sub-problems are fused via a weighted average method. Numerical simulations are carried out on a frame structure and a truss structure, respectively. A cluster constructed from three personal computers is used for implementation of the proposed method. Illustrated results show that the proposed method can be used for identifying the dynamic forces in long-time duration and saving the computing time. Some related issues are discussed as well.

关键词： parallel computing Force identification Sparse regularization Inverse problem

来源：评论

学校读者我要写书评

暂无评论

parallel computing of fuzzy integrals: Performance and test

引用

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS 2021年第2期41卷 3137-3159页

作者： Wang, Jinfeng Huang, Shuaihui Jiang, Fajian Zheng, Zhishen Ou, Jianbin Chen, Hao Chen, Runjian Wang, Wenzhong South China Agr Univ Coll Math & Informat Guangzhou Peoples R China Guangzhou Key Lab Smart Agr Guangzhou Peoples R China Guangdong Elect Certificat Author Co LTD Guangzhou Peoples R China South China Agr Univ Coll Econ & Management Guangzhou Peoples R China

Fuzzy integral in data mining is an excellent information fusion tool. It has obvious advantages in solving the combination of features and has more successful applications in classification problems. However, with the increase of the number of features, the time complexity and space complexity of fuzzy integral will also increase exponentially. This problem limits the development of fuzzy integral. This article proposes a high-efficiency fuzzy integral-parallel and Sparse Frame Based Fuzzy Integral (PSFI) for reducing time complexity and space complexity in the calculation of fuzzy integrals, which is based on the distributed parallel computing framework-Spark combined with the concept of sparse storage Aiming at the efficiency problem of the Python language, Cython programming technology is introduced in the meanwhile. Our algorithm is packaged into an algorithm library to realize a more efficient PSFI. The experiments verified the impact of the number of parallel nodes on the performance of the algorithm, test the performance of PSFI in classification, and apply PSFI on regression problems and imbalanced big data classification. The results have shown that PSFI reduces the variable storage space requirements of datasets with aplenty of features by thousands of times with the increase of computing resources. Furthermore, it is proved that PSFI has higher prediction accuracy than the classic fuzzy integral running on a single processor.

关键词： parallel computing sparse storage fuzzy integral fuzzy measure

来源：评论

学校读者我要写书评

暂无评论

parallel computing Method of Commonly Used Interpolation Algorithms for Remote Sensing Images

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2024年 17卷 315-322页

作者： Fan, Minghu Zuo, Xianyu Zhou, Bing Henan Univ Henan Key Lab Big Data Anal & Proc Kaifeng 475004 Peoples R China Henan Univ Sch Comp & Informat Engn Kaifeng 475004 Peoples R China

parallel computing is a common method to accelerate remote sensing image processing. This article briefly describes six commonly used interpolation functions and studies three commonly used parallel computing methods of the corresponding nine interpolation algorithms in remote sensing image processing. First, two kinds of general parallel interpolation algorithms (for CPU and GPU, respectively) are designed. Then, in two typical application scenarios (data-intensive and computing-intensive), four computing methods (one serial method and three parallel methods) of these interpolation algorithms are tested. Finally, the acceleration effects of all parallel algorithms are compared and analyzed. On the whole, the acceleration effect of the parallel interpolation algorithm is better in computer-intensive scenario. In CPU-oriented methods, the speedup of all parallel interpolation algorithms mainly depends on the number of physical cores of CPU, whereas in GPU-oriented methods, a speedup is greatly affected by the computation complexity of an algorithm and the application scenario. GPU has a better acceleration effect on the interpolation algorithms with bigger computation complexity and has more advantages in the computing-intensive scenarios. In most cases, GPU-based interpolation is ideal for efficient interpolation.

关键词： Compute-intensive data-intensive interpolation parallel computing remote sensing image

来源：评论

学校读者我要写书评

暂无评论

parallel computing method of deep belief networks and its application to traffic flow prediction

引用

KNOWLEDGE-BASED SYSTEMS 2019年 163卷 972-987页

作者： Zhao, Lu Zhou, Yonghua Lu, Huapu Fujita, Hamido Beijing Jiaotong Univ Sch Elect & Informat Engn Beijing Peoples R China Tsinghua Univ Inst Transportat Engn Beijing Peoples R China Iwate Prefectural Univ Fac Software & Informat Sci Takizawa Japan

Deep belief networks (DBNs) with outstanding advantages of learning input data features have attained particular attention and are applied widely in image processing, speech recognition, natural language interpretation, disease diagnosis, among others. However, owing to large data, the training processes of DBNs are time-consuming and may not satisfy the requirements of real-time application systems. In this study, a single dataset is decomposed into multiple subdatasets that are distributed to multiple computing nodes. Multiple computing nodes learn the features of their own subdatasets. On the precondition of the remaining features where one computing node learns from the total dataset, the single dataset learning models and algorithms are extended to the cases where multiple computing nodes learn multiple subdatasets in a parallel manner. Learning models and algorithms are proposed for the parallel computing of DBN learning processes. A master-slave parallel computing structure is designed, where the slave computing nodes learn the features of their respective subdatasets and transmit them to the master computing node. The master computing node is critical in synthesizing the learned features from the respective slave computing nodes. The broadcast, synchronization, and synthesis are repeated until all features of subdatasets have been learned. The proposed parallel computing method is applied to traffic flow prediction using practical traffic flow data. Our experimental results verify the effectiveness of the parallel computing method of DBN learning processes in terms of decreasing pre-training and fine-tuning times and maintaining the prominent feature learning abilities. (C) 2018 Elsevier B.V. All rights reserved.

关键词： Deep learning Deep belief network parallel computing Traffic flow prediction

来源：评论

学校读者我要写书评

暂无评论

parallel computing forty years ago

引用

MATHEMATICS AND COMPUTERS IN SIMULATION 1999年第1-2期51卷 47-62页

作者： Gustafson, K Univ Colorado Dept Math Boulder CO 80309 USA

The author's initiation into computing 40 years ago, and the early development of parallel computing paradigms within hardware and software systems which evolved at that time, are recounted in a personal trace. It was an optimal time to learn computing and the intuition gained has been of great value to the author ever since. It is hoped that the following account of one segment of early computing will be of interest especially to the younger readers. (C) 1999 IMACS/Elsevier Science B.V. All rights reserved.

关键词： parallel computing LGP-30 system D-825 computer

来源：评论

学校读者我要写书评

暂无评论

parallel computing and SGD-Based DPMM For Soft Sensor Development With Large-Scale Semisupervised Data

引用

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS 2019年第8期66卷 6362-6373页

作者： Shao, Weiming Yao, Le Ge, Zhiqiang Song, Zhihuan Zhejiang Univ State Key Lab Ind Control Technol Hangzhou 310007 Zhejiang Peoples R China Zhejiang Univ Coll Control Sci & Engn Hangzhou 310007 Zhejiang Peoples R China

Soft sensors based on Gaussian mixture models (GMM) have been widely used in industrial process systems for modeling the nonlinearity, non-Gaussianity, and uncertainties. However. there are still some challenging issues in developing high-accuracy GMM-based soft sensors. First, labeled samples are usually scarce due to technical or economical limitations, causing traditional supervised GMM-based soft sensing methods fail to provide satisfactory performance. Second, tremendous amounts of unlabeled samples are gathered, nevertheless, how to fully exploit those unlabeled samples in terms of improving both the predictive accuracy and computational efficiency remains unresolved. In this paper, in order to deal with these issues, two computationally efficient soft sensing methods, namely the parallel computing-based semisupervised Dirichlet process mixture models (P-S-2 DPMM) and stochastic gradient descent-based S-2 DPMM (SGD-S-2 DPMM), are proposed. The (SDPMM)-D-2 is first developed to mine information contained in both labeled and unlabeled samples for predictive accuracy enhancement, and subsequently is extended to the P-S-2 DPMM and SGD-S-2 DPMM to handle large-scale process data with sufficient and limited computing resources, respectively. Two case studies are carried out on real-world industrial processes, and the results obtained demonstrate the effectiveness of the proposed methods.

关键词： Dirichlet process mixture model (DPMM) parallel computing semisupervised learning soft sensor stochastic gradient descent (SGD)

来源：评论

学校读者我要写书评

暂无评论

parallel computing in finance for estimating risk-neutral densities through option prices

引用

JOURNAL OF parallel AND DISTRIBUTED computing 2023年 173卷 61-69页

作者： Monteiro, Ana M. Santos, Antonio A. F. Univ Coimbra Fac Econ CeBER Coimbra Portugal

Option pricing is one of the most active Financial Economics research fields. Black-Scholes-Merton option pricing theory states that risk-neutral density is lognormal. However, markets' pieces of evidence do not support that assumption. More realistic assumptions impose substantial computational burdens to calculate option pricing functions. Risk-neutral density is a pivotal element to price derivative assets, which can be estimated through nonparametric kernel methods. A significant computational challenge exists for determining optimal kernel bandwidths, addressed in this study through a parallel computing algorithm performed using Graphical Processing Units. The paper proposes a tailor-made Cross-Validation criterion function used to define optimal bandwidths. The selection of optimal bandwidths is crucial for nonparametric estimation and is also the most computationally intensive. We tested the developed algorithms through two data sets related to intraday data for VIX and S&P500 indexes.(c) 2022 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/).

关键词： Finance GPU computing Nonparametric methods Options pricing parallel computing

来源：评论

学校读者我要写书评

暂无评论

parallel computing of power flow for complex distribution network with DGs

引用

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS 2017年第5期33卷 2987-2997页

作者： Zhang, Limei Liu, Yongfu Zhang, Penghui Tang, Wei Lv, Jing Zhang, Lu Li, Dongming Agr Univ Hebei Coll Informat Sci & Technol Baoding Peoples R China China Agr Univ Coll Informat & Elect Engn Beijing Peoples R China Agr Univ Hebei Coll Mech & Elect Engn Baoding Peoples R China

With the penetration of multi-source and multi-type distributed generation (DG) in distribution network, the power flow calculations of distribution network become more and more complex. On the premise of analysis of power flow calculation model for different DG, this paper presents a power flow parallel computing algorithm for complex distribution network based on multicore CPU technique. On the basis of analysis of topological structure of distribution system, the task assignment problems in the isomorphism multicore processors were solved to improve speedup ratio and parallel efficiency. Integrated with the characteristics of backward/forward sweep methods, a hybrid power flow parallel computing method was given to adapt for the nodal type of various DG. In order to test the convergence and parallel efficiency, the proposed algorithm have been tested on IEEE 90 bus and a composite system, which is composed of IEEE 20 bus, IEEE 90 bus, IEEE 37 bus and etc. The results show that the proposed hybrid power flow computing method is adaptive to the complex distribution network with multi-type DGs, and the designed parallel algorithm effectively shortens time of solving equations and multicore resources are fully utilized.

关键词： Distributed generation distribution network power flow parallel computing

来源：评论

学校读者我要写书评

暂无评论

parallel computing Utilization in Nonlinear Model Predictive Control of Permanent Magnet Synchronous Motor

引用

IEEE ACCESS 2024年 12卷 128187-128200页

作者： Kozubik, Michal Vesely, Libor Aufderheide, Eyke Vaclavek, Pavel Brno Univ Technol Cent European Inst Technol CEITEC Brno 61200 Czech Republic Brno Univ Technol Dept Control & Instrumentat Brno 61600 Czech Republic Tech Univ Munich Chair Elect Dr Syst & Power Elect D-80333 Munich Germany

Permanent Magnet Synchronous Motor (PMSM) drives are widely used for motion control industrial applications and electrical vehicle powertrains, where they provide a good torque-to-weight ratio and a high dynamical performance. With the increasing usage of these machines, the demands on exploiting their abilities are also growing. Usual control techniques, such as field-oriented control (FOC), need some workaround to achieve the requested behavior, e.g., field-weakening, while keeping the constraints on the stator currents. Similarly, when applying the linear model predictive control, the linearization of the torque function and defined constraints lead to a loss of essential information and sub-optimal performance. That is the reason why the application of nonlinear theory is necessary. Nonlinear Model Predictive Control (NMPC) is a promising alternative to linear control methods. However, this approach has a major drawback in its computational demands. This paper presents a novel approach to the implementation of PMSMs' NMPC. The proposed controller utilizes the native parallelism of population-based optimization methods and the supreme performance of field-programmable gate arrays to solve the nonlinear optimization problem in the time necessary for proper motor control. The paper presents the verification of the algorithm's behavior both in simulation and laboratory experiments. The proposed controller's behavior is compared to the standard control technique of FOC and linear MPC. The achieved results prove the superior quality of control performed by NMPC in comparison with FOC and LMPC. The controller was able to follow the Maximal Torque Per Ampere strategy without any supplementary algorithm, altogether with constraint handling.

关键词： Torque parallel processing Predictive control Optimization Permanent magnet motors Vectors Stators Evolutionary algorithms motor control nonlinear control parallel computing predictive control

来源：评论

学校读者我要写书评

暂无评论

parallel computing of three-dimensional discontinuous deformation analysis based on OpenMP

引用

COMPUTERS AND GEOTECHNICS 2019年第Feb.期106卷 304-313页

作者： Peng, Xinyan Chen, Guangqi Yu, Pengcheng Zhang, Yingbin Guo, Longxiao Wang, Cungen Cheng, Xiao Niu, Hui Kyushu Univ Dept Civil & Struct Engn Fukuoka Fukuoka 8190395 Japan Southwest Jiaotong Univ Sch Civil Engn Dept Geotech Engn Chengdu 610031 Sichuan Peoples R China Southwest Jiaotong Univ Minist Educ Key Lab Transportat Tunnel Engn Chengdu 610031 Sichuan Peoples R China Northeastern Univ Sch Resources & Civil Engn Shenyang 110819 Liaoning Peoples R China Cent S Univ Sch Civil Engn Changsha 410075 Hunan Peoples R China

The computing efficiency of three-dimensional discontinuous deformation analysis (3D-DDA) needs to be improved for large-scale simulations. Among all the subroutines of 3D-DDA, the equation solver is very time-consuming. To accelerate the equation-solving process, this paper proposes implementing the parallel block Jacobi (BJ) and preconditioned conjugate gradient (PCG) iterative solvers into the original 3D-DDA based of OpenMP. The calculation accuracy and computational efficiency are studied by several numerical examples, demonstrating that the modified 3D-DDA with parallel BJ or PCG solver exhibits much higher execution effciency with satisfactory correctness. The maximum speedup ratio is up to 5.1 for the cases studied.

关键词： 3D-DDA Block Jacobi Preconditioned conjugate gradient OpenMP parallel computing

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：