EAGER: Leveraging Smart Meter Data for Enhanced Situational Awareness of Power Distribution Systems

由于消费者拥有的屋顶太阳能发电、电动汽车、储能和高级计量基础设施（AMI）的集成，配电系统变得越来越活跃。 建议的研究的目标是调查的基本方法，有效地整合AMI数据，以提高配电系统的态势感知更有效的操作。此外，该研究将专注于整合机器学习模型，从历史数据中学习，以提高建模准确性和不确定性量化。 该研究将通过展示AMI和其他智能电表数据对配电系统高效运行的价值，从根本上推进配电系统基于数据的运行。除了加速采用数据驱动的网格管理策略，拟议的工作可以作为一个实用的指南，公用事业利用AMI数据的态势感知。公用事业公司将能够更有效地操作系统，这反过来又会为公用事业公司和消费者带来经济利益。该项目将支持和培训一名博士生，使其掌握配电系统操作的基于数据的方法。 结果将通过会议和期刊出版物传播。该研究将调查智能电表的最佳轮询的创新方法，以获得低压配电系统的整体画面，协调多时间尺度的异构电表数据，以增强新的基于稀疏性的状态估计策略，并通过包括不确定性量化的机器学习框架，系统地整合历史数据和真实的时间测量。所提出的数据驱动的方法，利用真实的时间测量从一个有限的子集的智能电表代表了一个显着的范式转变，在当前的操作实践。该奖项反映了NSF的法定使命，通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Electric power distribution systems are becoming active due to integration of consumer owned rooftop solar generation, electric vehicles, energy storage, and advanced metering infrastructure (AMI). The goal of the proposed research is to investigate fundamental approaches to effectively integrate AMI data to increase situational awareness of distribution systems for more effective operation. Further, the research will focus on integrating machine learning models to learn from historical data for increasing modelling accuracy and uncertainty quantification. The research will result in fundamental advancement in data-based operation of power distribution systems by demonstrating the value of AMI and other smart meter data for efficient operation of the distribution system. In addition to accelerating the adoption of data-driven grid management strategies, the proposed work can serve as a practical guide for utilities to leverage AMI data for situational awareness. The utilities will be able operate the system more efficiently which in turn will lead to economic benefits for the utilities and the consumer. The project will support and train a doctoral student on data-based approaches for operation of power distribution systems. The results will be disseminated through conference and journal publications. The research will investigate innovative approaches for optimal polling of smart meter to obtain a holistic picture of the low-voltage distribution system, reconciling multi-time scale heterogenous meter data to empower novel sparsity-based state estimation strategies, and systematic integration of historical data and real time measurements via a machine learning framework that includes uncertainty quantification. The proposed data-driven approach that leverages real time measurements from a limited subset of smart meters represents a significant paradigm shift in current operational practices. It will also lead to new pathways for future research on utilizing the AMI data to enhance reliability, resiliency, and efficiency of power distribution systems.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.