检索结果-内蒙古大学图书馆

38th Chinese Control Conference (CCC)

作者： Zou, Yao Huang, Bomin Meng, Ziyang Ren, Wei Univ Sci & Technol Beijing Sch Automat & Elect Engn Beijing 100083 Peoples R China Tsinghua Univ Dept Precis Instrument Beijing 100084 Peoples R China Univ Calif Riverside Dept Elect & Comp Engn Riverside CA 92521 USA

ISBN: (纸本)9789881563972

This paper studies the Nash equilibrium seeking problem for two-layer constrained non-cooperative games. In particular, the non-cooperative game among upper-layer clusters is played by lower-layer players therein subject to set constraints. In each cluster, a player is appointed as an administrator who is responsible for interacting with other clusters. A KKT condition is first determined to formulate the underlying generalized Nash equilibrium. Based on such a KKT condition, a distributed seeking algorithm using a projection operator is proposed. In particular, under the assumption that each player has access to the administrators in other clusters, the proposed seeking algorithm via intra-cluster interaction guarantees that each player is capable of seeking its corresponding component of the generalized Nash equilibrium. An illustrative example effectively verifies the proposed theoretical results.

关键词： Non-cooperative game Nash equilibrium seeking distributed algorithm distributed observer

来源：评论

学校读者我要写书评

暂无评论

Coordinated optimization of P2P Energy Trading and Network Operation for Active Distribution Network with Multi-Microgrids

引用

Global Energy Interconnection 2025年

作者： Peishuai LI Yihan WANG Tao ZHENG Yulong JIN Weizhi YUAN Wenwen GUO State Key Laboratory of Technology and Equipment for Defense against Power System Operational Risks Nari Technology Co. Ltd. Nanjing China Nanjing University of Science&Technology Nanjing China

Microgrids (MGs) and active distribution networks (ADNs) are important platforms for distributed energy resource (DER) consumption. The increasing penetration of DERs has motivated the development ADNs coupled with MGs. This paper proposes a distributed co-optimization method for peer-to-peer (P2P) energy trading and network operation for an ADN integrated with multiple microgrids (MMGs). A framework that optimizes P2P energy trading among MMGs and ADN operations was first established. Subsequently, an energy management model that aims to minimize the operation and energy trading costs was constructed for each MG. Accordingly, the MMGs’ cooperative game model was established based on Nash bargaining theory to incentivize each stakeholder to participate in P2P energy trading, and a distributed solution method based on the alternating direction method of multipliers was developed. Moreover, an algorithm that adjusts the amount of energy trading between the ADN and MG is proposed to ensure safe operation of the distribution network. With the communication between the MG and ADN, the MMGs’ P2P trading and ADN operations are optimized in a coordinated manner. Finally, numerical simulations were conducted to verify the accuracy and effectiveness of the proposed method.

关键词： peer-to-peer energy trading game theory distributed algorithm distribution network operation

来源：评论

学校读者我要写书评

暂无评论

HPDL: Towards a General Framework for High-performance distributed Deep Learning 39

HPDL: Towards a General Framework for High-performance Distr...

引用

39th IEEE International Conference on distributed Computing Systems (ICDCS)

作者： Li, Dongsheng Lai, Zhiquan Ge, Keshi Zhang, Yiming Zhang, Zhaoning Sun, Tao Wang, Qinglin Wang, Huaimin Natl Univ Def Technol Sch Comp Natl Key Lab Parallel & Distributed Proc Changsha Peoples R China

ISBN: (纸本)9781728125190

With growing scale of the data volume and neural network size, we have come into the era of distributed deep learning. High-performance training and inference on distributed computing systems has been attracting increasing research attention in both academia and industry. Meanwhile, diversity of existing machine learning frameworks (e.g. TensorFlow, Pytorch and MXNet) and the explosion of deep learning hardwares (e.g. CPUs, GPUs, FPGAs and ASICs) bring more challenges for users to leverage new deep learning technologies and accelerating capability of hardware devices. We firstly search around the state-of-the-art work in the area which open our mind to take a vision upon the future deep learning framework. Then, we propose HPDL, a general framework for high-performance distributed deep learning which is compatible with existing frameworks and adaptive to various hardware architectures. At last, we discuss and foresee the key technologies fulfilling high-performance and large-scale deep learning, including optimization algorithm, hybrid communication mechanism, model parallelization, resource scheduling and single-node execution optimization.

关键词： deep learning distributed algorithm high-performance framework

来源：评论

学校读者我要写书评

暂无评论

distributed Consensus for Smart Grid Economic Dispatch with Prohibited Operating Zones 16

Distributed Consensus for Smart Grid Economic Dispatch with ...

引用

16th International Multi-Conference on Systems, Signals and Devices (SSD)

作者： Hamdi, Mounira Idomghar, Lhasanne Chaoui, Mondher Kachouri, Abdennaceur Natl Engn Sch Sfax LETI Lab ENIS BPW Sfax 3038 Tunisia Univ Haute Alsace IRIMAS Inst F-68093 Mulhouse France

ISBN: (纸本)9781728118208

The present paper exhibits a fully distributed cooperative algorithm based on consensus protocols to solve the economic dispatch problem (EDP) in the concept of the smart grids. In particular, the proposed protocol eliminates the need for a leader or a coordinated center and considers the vulnerability of the communication networks such as the link or/and node failure. In the power layer, for the formulation of the EDP, we consider the power generation limit, the ramp-rate limits, the power balance supply and the prohibited operating zones constraints, which makes the solution space of the (EDP) discontinuous. The evaluation of the proposed algorithm is effected on the IEEE 30 bus test system. Compared to other existing algorithms, numerical results prove the efficiency of the proposed approach to solve the discontinuous decentralized EDP under a practical communication network

关键词： component Smart Grid distributed economic dispatch prohibited operating zones consensus protocol distributed algorithm random communication topology

来源：评论

学校读者我要写书评

暂无评论

Building Temperature Control: A distributed Escort Dynamical approach 9

Building Temperature Control: A Distributed Escort Dynamical...

引用

9th International Conference on Power and Energy Systems (ICPES)

作者： Sawant, M. Moyalan, J. Koonamparampath, J. Sheikh, A. Wagh, S. Singh, N. VJTI EED Mumbai Maharashtra India Veermata Jijabai Technol Inst Elect Engn Dept Mumbai 400019 Maharashtra India

ISBN: (纸本)9781728126586

The constrained multi-agent optimization problem of building temperature control (BTC) is addressed in the framework of distributed resource allocation (DRA) using evolutionary game theory (EGT). The relationship between limited resource availability of DRA problem and constant population size requirement is established to devise population dynamic based control law. The BTC plant dynamics interconnected with the control dynamics are evaluated in the context of convergence to the rest point in terms of output consensus for dynamical systems. Furthermore, mitigating the expectation dependency of escort dynamical model of EGT, a DRA protocol is proposed referred to as distributed escort dynamics (DED). It ensures the satisfaction of global resource bound along with the local resource bounds on individual control parameters which is not the case for general dynamics such as replicator dynamics, logit dynamics, etc. referred to as classical models. The control action devised adopting DED approach is shown to provide smooth trajectory tracking along with the low start-up transience when compared with distributed interior point (DIP) method.

关键词： Consensus problems distributed algorithm Escort dynamics Graph theory Resource allocation

来源：评论

学校读者我要写书评

暂无评论

A distributed Approach to the Evasion Problem

引用

algorithmS 2020年第6期13卷 149-149页

作者： Khryashchev, Denis Chu, Jie Vejdemo-Johansson, Mikael Ji, Ping CUNY Grad Ctr Comp Sci 365 5th Ave New York NY 10016 USA CUNY Coll Staten Isl Math 2800 Victory Blvd Staten Isl NY 10314 USA CUNY John Jay Coll Math & Comp Sci 524 W 59th St New York NY 10019 USA

The Evasion Problem is the question of whether-given a collection of sensors and a particular movement pattern over time-it is possible to stay undetected within the domain over the same stretch of time. It has been studied using topological techniques since 2006-with sufficient conditions for non-existence of an Evasion Path provided by de Silva and Ghrist;sufficient and necessary conditions with extended sensor capabilities provided by Adams and Carlsson;and sufficient and necessary conditions using sheaf theory by Krishnan and Ghrist. In this paper, we propose three algorithms for the Evasion Problem: one distributed algorithm extension of Adams' approach for evasion path detection, and two different approaches to evasion path enumeration.

关键词： evasion path distributed algorithm

来源：评论

学校读者我要写书评

暂无评论

distributed Parareal in Time with Adaptive Coarse Solver for Large Scale Power System Simulations

Distributed Parareal in Time with Adaptive Coarse Solver for...

引用

IEEE-Power-and-Energy-Society General Meeting (PESGM)

作者： Osipov, Denis Duane, Nan Allu, Srikanth Simunovic, Srdjan Dimitrovski, Aleksandar Sun, Kai Univ Tennessee Dept Elect Engn & Comp Sci Knoxville TN 37996 USA Lawrence Livermore Natl Lab Computat Engn Div Livermore CA 94550 USA Oak Ridge Natl Lab Computat Sci & Engn Div Oak Ridge TN USA Univ Cent Florida Dept Elect & Comp Engn Orlando FL 32816 USA

ISBN: (纸本)9781728119816

This paper describes a distributed Parareal in time algorithm with the adaptive coarse solver. In the distributed algorithm, the propagation of a coarse solution is distributed among processors in addition to the fine solution propagation. In the adaptive coarse solver the complexity of the model is changed based on the size of generator rotor angle deviation. In addition, non-iterative approach to model a ZIP load is presented. The proposed method is tested on the large-scale 5617-machine 70285-bus Eastern Interconnection system.

关键词： Adaptive coarse solver coarse solution distributed algorithm fine solution Parareal in time

来源：评论

学校读者我要写书评

暂无评论

Segmentation and Abstraction of an IoT Enabled distributed Sensor Network 16

Segmentation and Abstraction of an IoT Enabled Distributed S...

引用

16th IEEE Annual Consumer Communications and Networking Conference (CCNC)

作者： Shao, Yuanhang Kumar, Suman Kawakami, Takahiro Troy Univ Dept Comp Sci Troy AL 36802 USA

ISBN: (纸本)9781538655535

We propose an area segmentation algorithm which is completely distributed, highly responsive, and exhibits a wide range of application scenarios. The proposed algorithm segments the area based on similarity of local sensor data and therefore, it requires a similarity measure parametrized with selected system indicators. In addition, algorithm creates an energy efficient data aggregation tree with a local highest energy node as a root. The resulting segmented sub-areas represents a level of spatial diversity and an abstraction of the sensor field which has a wide range of large scale distributed applications. Through simulation, the application and working of our scheme is demonstrated.

关键词： Sensor Networks Segmentation Abstraction Internet-Of-Things distributed algorithm

来源：评论

学校读者我要写书评

暂无评论

Cost and Latency Tradeoff in Mobile Edge Computing: A distributed Game Approach

Cost and Latency Tradeoff in Mobile Edge Computing: A Distri...

引用

IEEE International Conference on Big Data and Smart Computing (BigComp)

作者： Zaw, Chit Wutyee Ei, Nway Nway Im, Han Yeo Reum Tun, Yan Kyaw Hong, Choong Seon Kyung Hee Univ Dept Comp Sci & Engn Yongin South Korea

ISBN: (纸本)9781538677896

Resource allocation has been a critical issue in Mobile Edge Computing (MEC) since MEC servers colocated at Base Stations (BS) are limited in resources and users are competitive in nature. Users' devices have restraint on power and computation capability. Moreover, users want to offload their tasks to meet their latency deadlines and they have cost for offloading their tasks. In this paper, we analyze this tradeoff between cost and latency for offloading tasks. Basically, if users want to have less latency, they need more resources to accomplish this. But, more resources to allocate to users means they have to pay more cost. In addition, we formulate the resource allocation as Generalized Nash Equilibrium Problem because users' strategies are conflicted with one another when they compete for the resources from the resources pool at the BS. We propose a distributed algorithm for the game formulation since users prefer to control their own resources rather than revealing their information to others. Then, we analyze the Price of Anarchy numerically.

关键词： Mobile Edge Computing Resource Allocation Generalized Nash Equilibrium Problem distributed algorithm

来源：评论

学校读者我要写书评

暂无评论

distributed Tensor Decomposition for Large Scale Health Analytics 19

Distributed Tensor Decomposition for Large Scale Health Anal...

引用

World Wide Web Conference (WWW)

作者： He, Huan Henderson, Jette Ho, Joyce C. Emory Univ Atlanta GA 30322 USA CognitiveScale Austin TX USA

ISBN: (纸本)9781450366748

In the past few decades, there has been rapid growth in quantity and variety of healthcare data. These large sets of data are usually high dimensional (e.g. patients, their diagnoses, and medications to treat their diagnoses) and cannot be adequately represented as matrices. Thus, many existing algorithms can not analyze them. To accommodate these high dimensional data, tensor factorization, which can be viewed as a higher-order extension of methods like PCA, has attracted much attention and emerged as a promising solution. However, tensor factorization is a computationally expensive task, and existing methods developed to factor large tensors are not flexible enough for real-world situations. To address this scaling problem more efficiently, we introduce SG ran ite, a distributed, scalable, and sparse tensor factorization method fit through stochastic gradient descent. SG ran ite offers three contributions: (1) Scalability: it employs a block partitioning and parallel processing design and thus scales to large tensors, (2) Accuracy: we show that our method can achieve results faster without sacrificing the quality of the tensor decomposition, and (3) FlexibleConstraints: we show our approach can encompass various kinds of constraints including l(2) norm, l(1) norm, and logistic regularization. We demonstrate SGranite's capabilities in two real world use cases. In the first, we use Google searches for flu-like symptoms to characterize and predict influenza patterns. In the second, we use SGranite to extract clinically interesting sets (i.e., phenotypes) of patients from electronic health records. Through these case studies, we show SGranite has the potential to be used to rapidly characterize, predict, and manage a large multimodal datasets, thereby promising a novel, data-driven solution that can benefit very large segments of the population.

关键词： Web Mining User-Generated Content Health Analytics Tensor Decomposition distributed algorithm Apache Spark

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：