SCC-PG: Advanced Learning for Energy Risk Tracking (ALERT)

SCC-PG：能源风险跟踪高级学习 (ALERT)

• 批准号: 1951813
• 负责人: Mladen Kezunovic
• 金额: $ 15万
• 依托单位: Texas A&M Engineering Experiment Station
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2021-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1951813&HistoricalAwards=false
• 关键词: SCC; PG; Advanced; Learning; Energy

The energy supply is the backbone of Smart and Connected Communities (S&CC). It ties together various energy stakeholders (providers, consumers, and services) across different social, economic and cultural layers, and among groups corresponding to the residential, commercial, or industrial settings. Recent statistics indicate that despite all the measures taken by the utility industry to maintain the reliability and security of the energy supply, the number of major blackouts throughout the world is increasing. Loss of electricity affects residents, students, private and public sector employees, small and large businesses, and critical city services, such as, police, firefighters, and first responders. The fundamental question is whether the data sciences and engineering, combined with social sciences and technology can help reduce the losses and related societal impacts. This project postulates that the combination of the Big Data spatiotemporal analytics and physical models will allow us to achieve predictive outage risk capabilities to address the mitigation options not presently available. The Advance Learning for Energy Risk Tracking (ALERT) approach will focus on predicting the outages and asset failure risks using historical data from utilities (outage, smart meter data, etc.), along with additional data from weather-related government and private sources (radar, satellite, ASOS, NLDN and NDFD, Vaisala), as well as topology and vegetation data. Such risk predictions will be shared with participating utility companies, and with their customers to mobilize mitigation measures, which needs a strong social study aspect to better understand customer behavior. Those measures may include: a) equipment repair/replacement and feeder switching actions, b) relocation of the volatile population, and local power back up for schools, businesses and essential city services, c) scheduling firefighter and emergency services for evacuation of at-risk populations, and d) dispatching police forces to prevent looting of vacated houses and businesses. The goal of the communities to minimize the risk of electricity supply failure and undesirable environmental impacts will be achieved by engaging community stakeholders in creating and sharing outage prediction risk maps to allow for individual and collaborative mitigation actions. The main objective is to build research capability to develop a methodology for predicting the risk of electricity outages, which emphasizes the S&CC aspects. To strengthen the community engagement, and to consolidate thinking in the research team and among the various stakeholder groups in metropolitan areas in Texas, different types of meetings will be held.. These multidisciplinary meetings will focus on defining ALERT goals: the need for data collection and behavioral pilots, the requirements for the integration platform, the logistics of the user portal implementations for the dissemination of ALERT messages and handling of false positive and false negative scenarios, as well as collection of panel polling data in future steps. The project will identify the research gaps and means of addressing them, the data requirements, and will develop a comprehensive network of contacts for each partner organization, and each prospective pilot participant.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

能源供应是智能互联社区 (S&CC) 的支柱。它将不同社会、经济和文化层面以及与住宅、商业或工业环境相对应的群体之间的各种能源利益相关者（供应商、消费者和服务）联系在一起。最近的统计数据表明，尽管公用事业行业采取了所有措施来维持能源供应的可靠性和安全性，但世界各地的大停电次数仍在增加。断电会影响居民、学生、私营和公共部门员工、小型和大型企业以及重要的城市服务部门，例如警察、消防员和急救人员。根本问题是数据科学和工程与社会科学技术相结合是否可以帮助减少损失和相关的社会影响。该项目假设大数据时空分析和物理模型的结合将使我们能够实现预测中断风险的能力，以解决目前不可用的缓解选项。能源风险跟踪高级学习 (ALERT) 方法将侧重于使用公用事业公司的历史数据（停电、智能电表数据等）以及来自天气相关政府和私人来源的附加数据来预测停电和资产故​​障风险（雷达、卫星、ASOS、NLDN 和 NDFD、维萨拉）以及拓扑和植被数据。 此类风险预测将与参与的公用事业公司及其客户共享，以动员缓解措施，这需要强大的社会研究方面才能更好地了解客户行为。这些措施可能包括：a) 设备维修/更换和馈线切换行动，b) 重新安置不稳定的人口，以及为学校、企业和基本城市服务提供当地电力备份，c) 安排消防员和紧急服务人员疏散现场人员危险人群，以及 d) 派遣警察部队，防止对腾出的房屋和企业进行抢劫。通过让社区利益相关者参与创建和共享停电预测风险图，以采取单独和协作的缓解行动，可以实现社区最大限度地减少电力供应故障风险和不良环境影响的目标。主要目标是建立研究能力，开发预测停电风险的方法，该方法强调 S&CC 方面。为了加强社区参与，并巩固研究团队和德克萨斯州大都市区各个利益相关群体的思想，将举行不同类型的会议。这些多学科会议将重点关注确定警报目标：数据收集的必要性和行为试点、集成平台的要求、用于​​传播警报消息的用户门户实施的物流和处理误报和漏报情况，以及未来步骤中小组投票数据的收集。该项目将确定研究差距和解决这些差距的方法、数据要求，并将为每个合作伙伴组织和每个潜在的试点参与者建立一个全面的联系网络。该奖项反映了 NSF 的法定使命，并被认为值得通过以下方式获得支持：使用基金会的智力价值和更广泛的影响审查标准进行评估。