Due to the complexity of the structure of doubly-fed induction generator (DFIG), it is difficult to conduct real-time simulation. For this reason, this paper designs an embeddable parallel digital image IP of DFIG bas...
详细信息
Due to the complexity of the structure of doubly-fed induction generator (DFIG), it is difficult to conduct real-time simulation. For this reason, this paper designs an embeddable parallel digital image IP of DFIG based on field programmable logic array (FPGA) to realize efficient electromagnetic simulation of DFIG. First, this paper proposes a virtual capacitor equivalent method for induction machine equivalent "T" circuit decoupling;Secondly, based on the principle of data independence and parallelism within the time step, a parallel algorithm for the internal components of DFIG is proposed;Finally, the FPGA-DFIG digital image constructed in this paper is connected to the RT-LAB/power grid, and hardware experiments are carried out under three working conditions: steady state, voltage sag and DC side fault. The experimental results are compared with the simulation results of DFIG grid connected system built in MATLAB/Simulink environment to verify the effectiveness and rapidity of the methods and models proposed in this paper.(C) 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of the 2nd international Joint conference on Energy and Environmental Engineering, CoEEE, 2022.
Nowadays, performance in HPC applications focuses on MPI efficiency as the de facto message-passing library to exploit parallelism. Features such as multithread and communication and processing overlap are continuousl...
详细信息
ISBN:
(纸本)9798350363074;9798350363081
Nowadays, performance in HPC applications focuses on MPI efficiency as the de facto message-passing library to exploit parallelism. Features such as multithread and communication and processing overlap are continuously studied to adapt to new platforms and a more significant number of processing units like GPU platforms. In this sense, recently, the MPI-4.0 standard introduced the partitioned point-to-point communication primitives to potentiate computation and communication overlapping. This paper introduces an innovative extension to MPI, specifically addressing partitioned communication for MPI-reduction primitives. Traditional reduction tasks conventionally involve processing the complete input vector following the conclusion of GPU computations. In contrast, our proposed methodology exploits message partitioning to process reduction tasks in real-time incrementally. This approach allows the system to process individual partitions of the input vector as they become available, removing the necessity to await the full completion of GPU computations before initiating the reduction. Our results demonstrate promising benefits, particularly for large message sizes. However, it is essential to acknowledge that optimizations at synchronization points remain potential bottlenecks, requiring meticulous analysis and consideration.
With the rapid development of the information age, computer technology and network technology are more and more mature, and the application of Internet technology is more and more extensive. On the basis of the gradua...
详细信息
The proceedings contain 25 papers. The topics discussed include: graph memory neural network with adaptive message passing mechanism;parallelization of flow calculations for a passenger airplane benchmark model on a m...
ISBN:
(纸本)9798400716904
The proceedings contain 25 papers. The topics discussed include: graph memory neural network with adaptive message passing mechanism;parallelization of flow calculations for a passenger airplane benchmark model on a multi-GPU server;optimized implementation of SM2 algorithm based on mobile platforms;a set of resource scheduling methods based on new Sunway many-core processor;parallel optimization of plasma single particle simulation program based on Sunway Bluelight II supercomputer;regression-classification parallel prediction: an online learning optimization for job scheduling backfill algorithm;target ranging and transmission line trip probability evaluation method based on divide-and-conquer thought;and smart grid intrusion detection based on device fingerprint technology.
This paper realizes the high parallel hardware acceleration of machine learning and artificial neural networks through resource reuse. This method meets the requirements of the application side for network diversifica...
详细信息
The proceedings contain 13 papers. The topics discussed include: a study on the performance of distributed storage systems in edge computing environments;RESCAPE: a resource estimation system for microservices with gr...
ISBN:
(纸本)9798350387339
The proceedings contain 13 papers. The topics discussed include: a study on the performance of distributed storage systems in edge computing environments;RESCAPE: a resource estimation system for microservices with graph neural network and profile engine;PrometheusMigrate: efficient live migration of confidential virtual machine with software abstraction;the cost perspective of adopting large language model-as-a-service;DCSA: the deployment mechanism of chained serverless applications in JointCloud environment;parallel computation in dynamic fog computingnetworks: a multi-armed bandit learning-based decentralized matching approach;and IBRI: an IoT solution for building collapse risk identification in smart cities.
Graph-based or multidimensional blockchains have been proposed to improve the scalability and efficiency of existing blockchain applications. However, when implemented in mobile Internet of Things (mIoT) networks, the...
详细信息
ISBN:
(纸本)9798350369458;9798350369441
Graph-based or multidimensional blockchains have been proposed to improve the scalability and efficiency of existing blockchain applications. However, when implemented in mobile Internet of Things (mIoT) networks, these blockchain systems can frequently split and merge and cause the merging algorithm to process a large number of similar blocks (a block is similar or identical when it exists once in multiple blockchains). The presence of similar blocks hinders the merging process, as it consumes time and computational resources to scan, validate, and potentially merge similar blocks. This paper presents an efficient approach for merging graph-based or multidimensional blockchains in mIoT networks by avoiding similar blocks and effectively detecting and merging new blocks that were created after the split. Our proposed merging algorithm employs depth-first search and Merkle tree techniques to minimize the time and computational resources spent on identical blocks. Finally, we evaluate the performance of our method in highly mobile networks and demonstrate that it can execute the merge with a more than 72% reduction in time in comparison to merging algorithms without block similarity handling.
Cellular automata have ideal properties for scalable and efficient computing, but a lack of a training method limits their real-world applications. First, we propose to partition the cellular lattice into three region...
详细信息
ISBN:
(纸本)9798350359329;9798350359312
Cellular automata have ideal properties for scalable and efficient computing, but a lack of a training method limits their real-world applications. First, we propose to partition the cellular lattice into three regions: input, output, and processing. Second, we propose a novel synthesis method to train a linear hybrid cellular automaton. Third, we show image classification on the MNIST dataset using only logic operations. By mapping local states over the globally linear lattice, the proposed model achieved above 90% test accuracy in binary image classification. Our method does not require any pre or post-processors to perform computation over the lattice. Hence, the lattice maintains its massive parallelism and locality of computation, ideal for ultra-low power processing in machine learning.
Using convolutional neural network technology for model compression, the convolutional layer operation speed is improved through multiple row-column parallelcomputing units in the convolution operation acceleration m...
详细信息
In recent years, in order to facilitate the efficient application of deep convolutional neural networks, it has become increasingly important to accelerate the inference stage of deep convolutional neural networks. Bu...
详细信息
暂无评论