Optimizing Energy Management in Microgrids with Datacenters: An Integrated Approach

优化数据中心微电网的能源管理：综合方法

基本信息

批准号：
1610471
负责人：
Shaolei Ren
金额：
$ 30万
依托单位：
University of California-Riverside
依托单位国家：
美国
项目类别：
Standard Grant
财政年份：
2016
资助国家：
美国
起止时间：
2016-07-15 至 2020-06-30
项目状态：
已结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1610471&HistoricalAwards=false
关键词：
Optimizing Energy Management Microgrids Datacenters

项目摘要

Microgrids are localized grids that have become an integral component of future smart grids. Meanwhile, datacenters have been expanding enormously to support the exploding digital economy worldwide, emerging as significant energy consumers that are commonly co-located with a diverse set of non-datacenter loads in microgrids. Nonetheless, prior research on datacenter energy management, although encouraging, has traditionally viewed datacenters as isolated stand-alone facilities and rarely considered their physical interconnection with other loads. On the other hand, the rich literature on microgrid energy management has primarily focused on non-datacenter loads (e.g., thermostatically controlled loads), while treating energy-intensive datacenters as miscellaneous loads and ignoring their unique global-routing capabilities. This lack of integration and coordination between datacenter and non-datacenter loads poses a series of challenges to microgrid management, such as high peak demand, waste of on-site renewables, and even potential threats to the stability of main grids. This project seeks to address the challenges and optimize microgrid energy management for minimizing the energy cost and facilitating demand responses to better protect the main grid. Towards this end, this project takes a transformative shift from the current view that treats datacenter and non-datacenter loads separately, to an integrated approach that holistically coordinates both datacenter and non-datacenter loads in microgrids. Specifically, this project investigates two complementary research thrusts. First, when the microgrid operator can directly schedule the loads, this project investigates novel control algorithms to holistically manage both datacenter and non-datacenter loads, while taking into account datacenters spatial-routing and heterogeneous communications capabilities. Second, when the loads are managed by self-interested entities, this project studies market mechanisms to coordinate both datacenter and non-datacenter loads for microgrid-level efficiency.This project advances the existing research on microgrid energy management by transforming datacenters' role in microgrids from miscellaneous loads into a valuable asset with high scheduling flexibilities. It can catalyze a shift in the way that microgrids evolve, bearing great economic, environmental and societal impacts. This project will also incorporate the research into existing courses and provide abundant opportunities to nurture and attract students, especially those from under-represented groups, to engage in research careers.

微电网是局部网格，已成为未来智能电网的组成部分。同时，数据中心一直在广泛扩展，以支持全球爆炸的数字经济，成为了重要的能源消费者，这些消费者通常与微电网中各种非dateCenter载荷共同列入。尽管如此，关于数据中心能源管理的事先研究尽管令人鼓舞，但传统上一直将数据中心视为孤立的独立设施，很少考虑其与其他负载的物理互连。另一方面，有关微电网能源管理的丰富文献主要集中在非围次载荷（例如，恒温控制的负载）上，同时将能源密集型数据中心视为其他载荷并忽略了其独特的全球透视功能。数据中心和非纳入人物负载之间缺乏整合和协调，给微电网管理带来了一系列挑战，例如高峰需求，浪费了现场可再生能源，甚至可能对主要网格稳定性的潜在威胁。该项目旨在应对挑战并优化微电网能源管理，以最大程度地降低能源成本并促进需求响应以更好地保护主要电网。为此，该项目从当前的视图分别处理数据中心和非传统载荷的当前视图转变为一种集成方法，该方法可以整体协调数据中心和非datacenter载荷。具体而言，该项目研究了两个互补的研究推力。首先，当微电网操作员可以直接安排负载时，该项目会研究新颖的控制算法，以整体管理数据中心和非纳入负载，同时考虑到数据中心的空间路由和异质通信功能。其次，当负载由自我利益实体管理时，该项目的市场机制可以协调微电流级效率的数据中心和非datacenter载荷。该项目通过将数据中心在微电流中的作用转化为有价值的柔韧性，通过将数据中心在微电流中的作用转化为微电流的作用，从而推进了微电能管理的现有研究。它可以催化微电网发展的方式，具有巨大的经济，环境和社会影响。该项目还将将研究纳入现有课程中，并为培育和吸引学生，尤其是代表性不足的学生从事研究职业的学生提供丰富的机会。

项目成果

期刊论文数量（8）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Increasing the Trustworthiness of Deep Neural Networks via Accuracy Monitoring

DOI：
发表时间：
2020-07
期刊：
ArXiv
影响因子：
0
作者：
Zhihui Shao;Jianyi Yang;Shaolei Ren
通讯作者：
Zhihui Shao;Jianyi Yang;Shaolei Ren

On the Latency Variability of Deep Neural Networks for Mobile Inference

关于移动推理深度神经网络的延迟可变性

DOI：
发表时间：
2020
期刊：
Workshop on Hot Topics in Edge Computing (1-page Poster
影响因子：
0
作者：
Yang, Luting;Lu, Bingqian;Ren, Shaolei
通讯作者：
Ren, Shaolei

Multi-Feedback Bandit Learning with Probabilistic Contexts

DOI：
10.24963/ijcai.2020/427
发表时间：
2020-07
期刊：
影响因子：
0
作者：
Da Yu;Huishuai Zhang;Wei Chen;Tie-Yan Liu;Jian Yin
通讯作者：
Da Yu;Huishuai Zhang;Wei Chen;Tie-Yan Liu;Jian Yin

Automating Deep Neural Network Model Selection for Edge Inference

DOI：
10.1109/cogmi48466.2019.00035
发表时间：
2019-12
期刊：
2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)
影响因子：
0
作者：
Bingqian Lu;Jianyi Yang;L. Chen;Shaolei Ren
通讯作者：
Bingqian Lu;Jianyi Yang;L. Chen;Shaolei Ren

DeepPM: Efficient Power Management in Edge Data Centers using Energy Storage