Collaborative Research: Data-Driven Situational Awareness for Resilient Operation of Distribution Networks with Inverter-based distributed energy resources

随着清洁、经济、可靠和有弹性的电力网络的重要性日益增加，分布式能源（DERs）的大规模集成，如太阳能发电和电池存储，已经被考虑并在配电网中实施。随着这些资源的激增，配电网和分布式电网集群的协调控制和管理对于在正常模式和电网突发事件下实现最优、可靠、弹性和稳定运行变得至关重要。这项研究将推进基于逆变器的分布式电网管理的科学基础，并通过释放来自住宅、商业或工业客户的额外电网服务能力，满足配电网络的新需求。该项目的成果预计将对大规模集成可再生逆变器的电网的可靠性和弹性产生重大影响。该合作提案的目标是将人工智能（AI）和机器学习与电力系统和电力电子概念相结合，设计新颖的态势感知和纠正行动识别工具，以实现具有高渗透率的基于逆变器的der的配电网络的可靠、弹性运行。本项目包括以下几个重点方面：1)开发了将配电网多速率时序数据集与基于逆变器的der集成的新方法，用于网络态势评估；2)建立一种强制的基于相干性的聚合，即使基于逆变器的聚类der最初是不相干的；3)开发基于稳定性的优化框架，用于电网突发事件中异构负载、临界负载和柔性负载的孤岛稳定集群边界识别。该项目的智力意义包括：1)开发了一种基于多速率、多传感器、基于概率图形模型的数据融合和配电网态势感知方法，该方法具有基于增殖型逆变器的DERs；2)建立基于相干性的聚合和动态模型开发技术，增强DER之间的相干性，实现DER集群的有效聚类，实现DER集群的真正聚合模型；3)开发基于Lyapunov稳定性的优化框架，用于配电网中异构负载、临界负载和柔性负载的自主孤岛集群边界识别，以提高电网在突发事件中的可靠性和弹性。该项目的成功完成将对基于逆变器的der大规模集成的电网可靠性和弹性产生重大影响。该奖项反映了美国国家科学基金会的法定使命，并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。

With the increasing importance of clean, affordable, reliable, and resilient electricity networks, large scale integration of distributed energy resources (DERs), such as solar generations, and battery storages, has been considered and implemented in distribution networks. With the proliferation of such resources, coordinated control and management of distribution networks and the DER clusters have become of upmost importance to achieve optimal, reliable, resilient, and stable operation during normal mode and grid contingencies. This research will advance the scientific foundations of grid management using inverter based DERs and allow the new requirements of distribution networks by unlocking additional grid services capabilities from residential, commercial or industrial customers. The outcome of this project is expected to have substantial impacts on reliability and resilience of our electricity network with large-scale integration of renewable inverter based DERs. The goal of this collaborative proposal is to combine artificial intelligence (AI) and machine learning with power systems and power electronics concepts to design novel situational awareness and corrective action identification tools for reliable, resilient operation of distribution networks with a high penetration of inverter-based DERs. This project includes the following key aspects; 1) Development of new methods to integrate multi-rate time-series data sets in distribution networks with inverter-based DERs for network situational assessment; 2) establishment of an enforced coherency-based aggregation for clustering inverter-based DERs even if they were initially non-coherent; and 3) development of a stability-based optimization framework for boundary identification of islanded stable clusters of heterogeneous DERs, critical, and flexible loads during grid contingencies. The intellectual significance of the project includes: 1) development of a multi-rate, multi-sensor, probabilistic graphical-model-based method for data fusion and distribution network situational awareness with proliferated inverter-based DERs; 2) establishment of a coherency-based aggregation and dynamic model development technique to enforce coherency and enable effective clustering among DERs and to realize a true aggregate model of DER clusters; and 3) development of a Lyapunov stability-based optimization framework for boundary identification of autonomously islanded clusters of heterogeneous DERs, critical and flexible loads in distribution networks to enhance grid reliability and resilience during grid contingencies. Successful completion of this project will have significant impacts on grid reliability and resilience with large-scale integration of inverter-based DERs.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.