检索结果-内蒙古大学图书馆

distributed Predefined-time Droop-Free Unified Control Scheme for parallel Energy Storage Units in All-Electric Aircraft Onboard DC Microgrid

引用

IEEE Transactions on Power Electronics 2025年

作者： Yu, Heyang Lin, Ping Zeng, Yuji Zhang, Qinjin Wu, Yuhu Dalian University of Technology Key Laboratory of Intelligent Control and Optimization for Industrial Equipment Ministry of Education School of Control Science and Engineering Dalian116024 China Dalian Maritime University College of Marine Engineering Dalian116026 China

In all-electric aircraft (AEA), onboard DC microgrids with parallel energy storage units (ESUs) often suffer from state-of-charge (SoC) imbalance, inaccurate current distribution, and DC bus voltage instability - threatening system safety and performance. This paper proposes a distributed droop-free unified control scheme based on predefined-time (PdT) control theory, addressing these challenges through two main components. First, a distributed PdT consensus observer allows each ESU to locally estimate global average values within user-defined timeframe by exchanging information only with neighboring units. Second, a PdT droop-free unified controller simultaneously achieves precise SoC-based current allocation and stable bus voltage regulation - without relying on droop coefficients or multiple secondary loops. Simulation results under communication delays, controller input noise, and network scalability conditions demonstrate that the proposed method restores bus voltage and synchronizes control variables within a predefined time. Compared to enhanced droop-based strategies, it reduces bus voltage fluctuation by up to 80.8%, and significantly improves SoC balancing speed and current distribution accuracy. Hardware-in-the-loop experiments on the Starsim platform validate the method's real-time performance and robustness under load transients and fault conditions. Overall, the proposed scheme offers a scalable, low-complexity solution for reliable coordination control in AEA microgrids with multiple ESUs. © 1986-2012 IEEE.

关键词： DC distribution systems

来源：评论

学校读者我要写书评

暂无评论

Simulating Heterogeneous Models on Multi-Core Platforms using Julia's Computing Language parallel Potential

Simulating Heterogeneous Models on Multi-Core Platforms usin...

引用

International workshop/Symposium on distributed Simulation and real-time Applications (DS-RT)

作者： Sergey Suslov Michael Schiek Markus Robens Christian Grewing Stefan van Waasen ZEA-2 - Electronic Systems Research Center Jülich Germany

ISBN: (数字)9781728173436

ISBN: (纸本)9781728173443

This paper addresses the questions of high-level system modelling using heterogeneous multi-tool modelling environment on parallel multi-core processing systems for simulation acceleration. The modelling technique has been applied for high-level validation of a high-precision Indoor Positioning System for Motion Analysis (IPS-MA) developed in the Central Institute Electronic systems (ZEA-2) of the Research Center Juelich GmbH. The heterogeneous modelling environment has been built using an implementation-level model designed in Matlab Simulink, a verification model for describing the system environment using Modelica language, and Julia language for automatic generation of binding modelling environment and parallelizing the simulation of the overall model. The approach showed a good flexibility in system description and verification in the multi-instrument modelling environment and a good performance gain due to simulation parallelism.

关键词： Handheld computers Computational modeling Software packages Patents Sensors

来源：评论

学校读者我要写书评

暂无评论

Online scheduling of heterogeneous distributed machine learning jobs 20

Online scheduling of heterogeneous distributed machine learn...

引用

21st ACM International Symposium on Mobile Ad Hoc Networking and Computing, MobiHoc 2020

作者： Zhang, Qin Zhou, Ruiting Wu, Chuan Jiao, Lei Li, Zongpeng Wuhan University China Chinese University of Hong Kong Hong Kong Hong Kong University of Hong Kong Hong Kong Hong Kong University of Oregon United States

ISBN: (纸本)9781450380157

distributed machine learning (ML) has played a key role in today's proliferation of AI services. A typical model of distributed ML is to partition training datasets over multiple worker nodes to update model parameters in parallel, adopting a parameter server architecture. ML training jobs are typically resource elastic, completed using various time lengths with different resource configurations. A fundamental problem in a distributed ML cluster is how to explore the demand elasticity of ML jobs and schedule them with different resource configurations, such that the utilization of resources is maximized and average job completion time is minimized. To address it, we propose an online scheduling algorithm to decide the execution time window, the number and the type of concurrent workers and parameter servers for each job upon its arrival, with a goal of minimizing the weighted average completion time. Our online algorithm consists of (i) an online scheduling framework that groups unprocessed ML training jobs into a batch iteratively, and (ii) a batch scheduling algorithm that configures each ML job to maximize the total weight of scheduled jobs in the current iteration. Our online algorithm guarantees a good parameterized competitive ratio with polynomial time complexity. Extensive evaluations using real-world data demonstrate that it outperforms state-of-the-art schedulers in today's AI cloud systems. © 2020 ACM.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

BEI-TAB: Enabling Secure and distributed Airport Baggage Tracking with Hybrid Blockchain-Edge System

BEI-TAB: Enabling Secure and Distributed Airport Baggage Tra...

引用

International Conference on Communication Technology (ICCT)

作者： Fei Wang Pengbo Si Enchang Sun Yuzhao Su Faculty of Information Technology Beijing University of Technology Beijing China

ISBN: (纸本)9781665432078

Global air transport carries about 7.3 billion pieces of luggage each year and up to 56 percent of travelers prefer obtaining real-time baggage tracking information throughout their trip. However, the traditional baggage tracking scheme is generally based on optical scanning and centralized storage systems, which suffers from low efficiency and information leakage. In this paper, a blockchain and edge computing based IOT system for tracking of airport baggage (BEI-TAB) is proposed. Through the combination of radio frequency identification technology (RFID) and blockchain, real-time baggage processing information is automatically stored in blockchain. In addition, we deploy Interplanetary File System (IPFS) at edge nodes with ciphertext policy attribute-based encryption (CP-ABE) to store basic baggage information. Only hash values returned by the IPFS network are kept in blockchain, enhancing the scalability of the system. Furthermore, a multi-channel scheme is designed to realize the physical isolation of data, and to rapidly process multiple types of data and business requirements in parallel. To the best of our knowledge, it is the first architecture that integrates RFID, IPFS, CP-ABE with blockchain technologies to facilitate secure, decentralized, and real-time characteristics for storing and sharing data for baggage tracking. We have deployed a testbed with both software and hardware to evaluate the proposed system, considering the performances of transaction processing time and speed.

关键词： Privacy Scalability Computer architecture Airports real-time systems Software Hardware

来源：评论

学校读者我要写书评

暂无评论

Fostering resiliency, the importance of the demand side

Fostering resiliency, the importance of the demand side

引用

20th International workshop on Large-Scale Integration of Wind Power into Power systems as well as on Transmission Networks for Offshore Wind Power Plants (WIW 2021)

作者： E. Bobbio

Price responsive demand is critical for resiliency. Texas 2021 winter storm resulted in a gap of 30GW between demand and supply and rolling outages for four days. If 80% of customers had been price responsive with a -0.1 elasticity, no outages would have been necessary (54% with a more realistic -0.2 elasticity). Widespread adoption of flexible contracts requires institutional and technical innovation. Contracts exposing the consumer to the real-time electricity price should have hedging. Hedging protects consumers from bill volatility while preserving incentives to use electricity based on social value (cut consumption when electricity is scarce and other people need to heat their houses). Technology will reduce cognitive costs and strengthen the price response. distributed generation and storage and investment in energy efficiency are crucial too. Electric vehicles might increase electricity demand by 50% by 2050 in the US, but their batteries could power the US economy for over two days. Flexible contracts are necessary for V2G deployment. Price responsive demand and V2G also support the energy transition by shifting demand to times of renewable generation abundance. Promoting the adoption of flexible contracts with hedging and the deployment of V2G is critical for a secure, sustainable, and least-cost energy transition.

关键词： battery powered vehicles contracts cost reduction demand side management distributed power generation energy conservation incentive schemes investment power generation economics power generation reliability power markets pricing renewable energy sources storms vehicle-to-grid

来源：评论

学校读者我要写书评

暂无评论

Hardware Accelerated Semantic Declarative Memory systems through CUDA and MapReduce

引用

IEEE TRANSACTIONS ON parallel AND distributed systems 2019年第3期30卷 601-614页

作者： Edmonds, Mark Atahary, Tanvir Douglass, Scott Taha, Tarek Univ Dayton Dept Elect & Comp Engn Dayton OH 45469 USA US Air Force Res Lab Dayton OH 45433 USA

Declarative memory enables cognitive agents to effectively store and retrieve factual memory in real-time. Increasing the capacity of a real-time agent's declarative memory increases an agent's ability to interact intelligently with its environment but requires a scalable retrieval system. This work represents an extension of the Accelerated Declarative Memory (ADM) system, referred to as Hardware Accelerated Declarative Memory (HADM), to execute retrievals on a GPU. HADM also presents improvements over ADM's CPU execution and considers critical behavior for indefinitely running declarative memories. The negative effects of a constant maximum associative strength are considered, and mitigating solutions are proposed. HADM utilizes a GPU to process the entire semantic network in parallel during retrievals, yielding significantly faster declarative retrievals. The resulting GPU-accelerated retrievals show an average speedup of approximately 70 times over the previous Service Oriented Architecture Declarative Memory (soaDM) implementation and an average speedup of approximately 5 times over ADM. HADM is the first GPU-accelerated declarative memory system in existence.

关键词： Declarative memory ACT-R semantic networks parallel activation calculation

来源：评论

学校读者我要写书评

暂无评论

A Personal distributed real-time Collaborative System

A Personal Distributed Real-time Collaborative System

引用

International Conference on parallel and distributed systems (ICPADS)

作者： Michalis Konstantopoulos Nikos Chondros Mema Roussopoulos University of Athens Greece

ISBN: (数字)9781728190747

ISBN: (纸本)9781728183824

In this paper, we present O 3 real, a privacy-preserving distributed middleware for real-time collaborative editing of documents. O 3 real introduces a novel approach for building peer-to-peer real-time collaborative applications, using a reliable broadcast channel mechanism for network communication, but at the same time provides for persistent storage management of collaborative documents using the filesystem interface of a POSIX compliant filesystem. This approach enables real-time, completely decentralized collaboration among users, without the need for a third party to intervene, and significantly simplifies the creation of peer-to-peer collaborative applications. We demonstrate that O 3 real scales well for real-time collaboration use-cases. For example, with 33 users simultaneously collaborating on a document in real time over a WAN with a 50 ms link delay, the average perceived latency is approximately 54 ms, which is very close to the optimal baseline. In comparison, Etherpad exhibits nearly twice the perceived latency.

关键词： Wide area networks Storage management Collaboration real-time systems Peer-to-peer computing Reliability Middleware

来源：评论

学校读者我要写书评

暂无评论

Implementation of parallel algorithm for automatic seam tracking system of petroleum steel pipe

Implementation of parallel algorithm for automatic seam trac...

引用

Information Science, parallel and distributed systems (ISPDS), International Conference on

作者： Xiaoling Yan Liming Wang Lichun Han College of electrical engineering Naval Engineering University Wuhan China

ISBN: (数字)9781728196688

ISBN: (纸本)9781728196695

The article presents a parallel processing method instead of traditional serial processing method used in the software of petroleum pipe welding seam tracking system. Firstly, it analyses the shortages of serial processing method, and then a parallel processing method is raised, at last, the software's working process is provided which uses parallel processing method. It is confirmed that this method saves data processing time and makes the whole system have good real time capability in practical project.

关键词： Welding real-time systems Information science

来源：评论

学校读者我要写书评

暂无评论

Rosella: A Self-Driving distributed Scheduler for Heterogeneous Clusters

Rosella: A Self-Driving Distributed Scheduler for Heterogene...

引用

International Conference on Mobile Ad-hoc and Sensor Networks, MSN

作者： Qiong Wu Zhenming Liu William & Mary

ISBN: (纸本)9781665406697

Large-scale interactive web services and advanced AI applications make sophisticated decisions in real-time, based on executing a massive amount of computation tasks on thousands of servers. Task schedulers, which often operate in heterogeneous and volatile environments, require high throughput, i.e., scheduling millions of tasks per second, and low latency, i.e., incurring minimal scheduling delays for millisecond-level tasks. Scheduling is further complicated by other users’ workloads in a shared system, other background activities, and the diverse hardware configurations inside *** present Rosella, a new self-driving, distributed approach for task scheduling in heterogeneous clusters. Rosella automatically learns the compute environment and adjusts its scheduling policy in real-time. The solution provides high throughput and low latency simultaneously because it runs in parallel on multiple machines with minimum coordination and only performs simple operations for each scheduling decision. Our learning module monitors total system load and uses the information to dynamically determine optimal estimation strategy for the backends’ compute-power. Rosella generalizes power-of-two-choice algorithms to handle heterogeneous workers, reducing the max queue length of O(logn) obtained by prior algorithms to O(logn). We evaluate Rosella with a variety of workloads on a 32-node AWS cluster. Experimental results show that Rosella significantly reduces task response time, and adapts to environment changes quickly.

关键词： Processor scheduling Web services Heuristic algorithms Clustering algorithms Throughput Dynamic scheduling real-time systems

来源：评论

学校读者我要写书评

暂无评论

distributed V2V Computing Offloading Method Based on Delay and Fairness Awareness

Distributed V2V Computing Offloading Method Based on Delay a...

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Guoliang Yang Mengqi Xiong Guangsheng Feng Yuzheng Liu Yun Huang College of Computer Science and Technology Harbin Engineering University Harbin China The Second Research Institute China Aerospace Science and Industry Corporation Beijing China

The roadside Basic unit (RSU) can relieve the heavy computing and analysis pressure required by vehicles’computing tasks. Due to the mobility of vehicles and limited computing resources, it is difficult to deal extensive vehicle tasks in time. Therefore, we consider the computation task offloading problem between the vehicles, as well as the dynamic topology caused by the vehicle movement and the randomness of task arrival, and propose an online offloading method based on fairness awareness, which can be ensured tolerable delays and the requirement of the quality of experience (QoE) between vehicles by introducing delay-aware and price-constrained virtual queues. By Lyapunov optimization technology, the objective function is transformed into a drift-plus-penalty profit minimum bound problem, achieveing all the offloading decisions by the realtime resource requirements and topology state. Finally, extensive simulations are performed to demonstrate the efficiency of the proposed algorithm.

关键词： Vehicular ad hoc networks Transforms Linear programming real-time systems Delays Topology Quality of experience

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：