PFI:BIC - Utility-driven Smart Energy Services

PFI：BIC - 公用事业驱动的智能能源服务

• 批准号: 1534080
• 负责人: Prashant Shenoy
• 金额: $ 100万
• 依托单位: University of Massachusetts Amherst
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-09-01 至 2019-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1534080&HistoricalAwards=false
• 关键词: PFI; BIC; Utility; driven; Smart

This Partnerships for Innovation: Building Innovation Capacity (PFI:BIC) project focuses on the development of a utility-driven energy service platform. Since buildings consume a large fraction of society's total energy usage, even modest improvements in building energy efficiency have the potential to yield significant benefits. In recent years, utilities have deployed tens of millions of smart electric meters that record building energy usage over short intervals (e.g., every few minutes). While the original purpose of smart meters was to support basic utility operations (e.g., automated meter reading), this project uses them as the foundation for developing a new class of smart energy service systems. Specifically, the project focuses on analyzing the vast amount of data available from utility smart meters and other networked sensors to improve the energy efficiency of buildings and the electric grid. The research is utility-driven, since utilities have i) access to massive amounts of customer energy data, ii) a way to deliver the results of analytics to customers, and iii) strong incentives to improve customer energy efficiency (e.g., by reducing peak demand to reduce their own operational costs). The approach is scalable, since it leverages already available building smart meter data, rather than requiring the installation of new smart devices and control systems. The project will develop utility-driven smart energy services to improve grid energy efficiency, encourage energy conservation, and promote the local renewable energy sources, such as rooftop solar. The research methodology includes developing new energy analytic techniques for smart meter data, combining these in novel ways to create specific applications that improve energy efficiency, packaging those applications as cloud-based web services that are accessible to end users, and evaluating their impact on user behavior and energy efficiency. In particular, the project will develop three specific smart energy service systems: i) iProgram to determine optimal thermostat schedules by analyzing smart meter data, ii) Green Demand Response (Green-DR) to opportunistically schedule elastic loads to align with renewable energy generation, and iii) PowerTrip to provide real-time energy event notifications and suggested conservation actions to users. The project will implement a prototype of each system and evaluate them on real-world data (rendered anonymous) from 18,000 smart meters. The research follows a human-in-the-loop model that integrates technical and user behavioral research in the design and implementation of smart energy services. Thus, the project includes user studies to assess each service's behavioral impact on users.Broader impacts stem from the project's industrial partnerships with a local utility, a premier research lab (Microsoft Research), a small startup company (Budderfly), and an environmentally-conscious non-profit (WattTime.org). These partnerships will expose students to industry perspectives on addressing real-world energy-efficiency problems, and interdisciplinary research that spans technical and behavioral disciplines. The project will enable sustainability-aware curriculum, undergraduate REU projects, and sustainability-focused outreach efforts to local K-12 students through the newly established Massachusetts Green High Performance Computing Center.The lead institution is the University of Massachusetts Amherst. The primary industry partner is Holyoke Gas and Electric (HG&E)) (Holyoke, MA), a municipal utility. Primary academic partners are the University of Maryland-Baltimore County (Baltimore, MD) and Williams College (Williamstown, MA). Broader context partners are Microsoft (Redmond, WA), a large business, WattTime.org (Berkeley, CA), a non-profit, and Budderfly (Shelton, CT), a small business.

该创新伙伴关系：建设创新能力 (PFI:BIC) 项目重点开发公用事业驱动的能源服务平台。由于建筑物消耗了社会总能源使用量的很大一部分，因此即使对建筑物能源效率进行适度的改进也有可能产生显着的效益。 近年来，公用事业公司部署了数千万个智能电表，这些电表可以在短时间内（例如每隔几分钟）记录建筑能源使用情况。 虽然智能电表的最初目的是支持基本的公用事业运营（例如自动抄表），但该项目将其用作开发新型智能能源服务系统的基础。 具体来说，该项目的重点是分析公用事业智能电表和其他网络传感器提供的大量数据，以提高建筑物和电网的能源效率。 该研究是公用事业驱动的，因为公用事业公司可以i）访问大量客户能源数据，ii）向客户提供分析结果的方式，以及iii）提高客户能源效率的强有力激励措施（例如，通过降低峰值）降低自身运营成本的需求）。 该方法具有可扩展性，因为它利用现有的建筑智能电表数据，而不需要安装新的智能设备和控制系统。该项目将开发公用事业驱动的智能能源服务，以提高电网能源效率，鼓励节约能源，并推广屋顶太阳能等当地可再生能源。研究方法包括为智能电表数据开发新的能源分析技术，以新颖的方式将这些技术结合起来，创建提高能源效率的特定应用程序，将这些应用程序打包为最终用户可以访问的基于云的 Web 服务，并评估它们对用户的影响行为和能源效率。 特别是，该项目将开发三个特定的智能能源服务系统：i）iProgram通过分析智能电表数据确定最佳恒温器时间表，ii）绿色需求响应（Green-DR）机会性地安排弹性负载以与可再生能源发电保持一致， iii) PowerTrip 向用户提供实时能源事件通知和建议的节能行动。 该项目将实现每个系统的原型，并根据来自 18,000 个智能电表的真实数据（匿名）对其进行评估。 该研究遵循人机交互模型，将技术和用户行为研究整合到智能能源服务的设计和实施中。因此，该项目包括用户研究，以评估每项服务对用户的行为影响。更广泛的影响源于该项目与当地公用事业公司、一流研究实验室（微软研究院）、一家小型初创公司（Budderfly）和一家环保公司的工业合作伙伴关系。有意识的非营利组织（WattTime.org）。 这些合作伙伴关系将使学生了解解决现实世界能源效率问题的行业视角，以及跨越技术和行为学科的跨学科研究。 该项目将通过新成立的马萨诸塞州绿色高性能计算中心，为当地 K-12 学生提供可持续发展意识课程、本科生 REU 项目以及以可持续发展为重点的外展活动。牵头机构是马萨诸塞大学阿默斯特分校。 主要行业合作伙伴是市政公用事业公司霍利奥克燃气电力公司 (HG&E)（马萨诸塞州霍利奥克）。 主要学术合作伙伴是马里兰大学巴尔的摩县分校（马里兰州巴尔的摩）和威廉姆斯学院（马萨诸塞州威廉斯敦）。 更广泛的合作伙伴包括大型企业 Microsoft（华盛顿州雷德蒙德）、非营利组织 WattTime.org（加利福尼亚州伯克利）和小型企业 Budderfly（康涅狄格州谢尔顿）。

项目成果

vSolar: Virtualizing Community Solar and Storage for Energy Sharing

vSolar：虚拟化社区太阳能和存储以实现能源共享

• DOI: 10.1145/3208903.3208932
• 发表时间: 2018-06
• 期刊: ACM International Conference on Future Energy Systems
• 作者: Lee, Stephen;Shenoy, Prashant;Ramamritham, Krithi;Irwin, David
• 通讯作者: Irwin, David

Weatherman: Exposing weather-based privacy threats in big energy data

气象员：揭露大能源数据中基于天气的隐私威胁

• DOI: 10.1109/bigdata.2017.8258032
• 发表时间: 2017-12
• 期刊: IEEE International Conference on Big Data
• 作者: Chen, Dong;Irwin, David
• 通讯作者: Irwin, David

On the Feasibility, Cost, and Carbon Emissions of Grid Defection

电网缺陷的可行性、成本和碳排放

• DOI: 10.1109/smartgridcomm.2019.8909691
• 发表时间: 2019-10
• 期刊: and Computing Technologies for Smart Grids (SmartGridComm
• 作者: Bansal, Akansha Singh;Irwin, David
• 通讯作者: Irwin, David

DeepRoof: A Data-driven Approach For Solar Potential Estimation Using Rooftop Imagery

DeepRoof：使用屋顶图像估算太阳能潜力的数据驱动方法

• DOI: 10.1145/3292500.3330741
• 发表时间: 2019-07
• 期刊: 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
• 作者: Lee, Stephen;Iyengar, Srinivasan;Feng, Menghong;Shenoy, Prashant;Maji, Subhransu
• 通讯作者: Maji, Subhransu

Helios: a programmable software-defined solar module

Helios：可编程软件定义太阳能模块

• DOI: 10.1145/3276774.3276783
• 发表时间: 2018-11
• 期刊: Proceedings of the 5th Conference on Systems for Built Environments
• 作者: Bashir, Noman;Irwin, David;Shenoy, Prashant
• 通讯作者: Shenoy, Prashant