CAREER: Scalable and Secure Control of Distributed Grid-Edge Resources for Enhanced Grid Reliability

可再生能源的日益普及给现代电力系统带来了前所未有的运行挑战，需要额外的电力系统灵活性来提高电网的可靠性。NSF CAREER项目旨在通过开发电网边缘资源（GER）管理框架来提高电力系统的灵活性，从而加强国家电网的脱碳和能源安全。该项目将通过克服三个相互交织的挑战，即高可扩展性要求、主要隐私问题和激增的网络安全风险，为GER管理带来革命性的变化。这将通过综合优化、机器学习、统计学和密码学的思想来实现，以建立一个高效、私有和安全的GER控制框架。该项目的智力优势包括：(1)开发一个可扩展的框架，使异构ger能够有效地协同控制；(2)研究强耦合去中心化GER控制的隐私保护措施；(3)研究针对分散GER控制算法的隐形网络攻击的模型、检测和缓解策略。该项目的更广泛影响包括：(1)释放异构ger以提高电网可靠性和深化电网脱碳；(2)提高电网以经济、安全、可靠的方式整合越来越多的可再生能源发电和ger的能力；(3)为行业提供开发新市场产品的洞察力。综合教育计划将向青少年、K-12学生和弱势群体传播控制和动力系统概念，激励他们追求STEM教育和职业。现有的GER控制框架可能会在大规模部署中失败，因为它们总是忽略由网络维度和GER异质性引起的可伸缩性问题，缺乏有效的措施来保护GER所有者的隐私，以及缺乏对隐蔽的有目的的网络攻击的理解。为了推进这方面的知识，该项目将(1)开发新的优化、机器学习和统计工具，以构建一个GER控制框架，该框架与GER类型无关，可根据GER人口和网络维度进行扩展，并适用于不同的网格服务；(2)表征GER管理中的隐私风险，并开发基于密码学和非基于密码学的隐私保护分散优化范式；(3)构成网络攻击向量，利用分散的GER控制算法实现隐形攻击目的，确定这些攻击的可检测性，并制定相应的检测和缓解策略。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。

The growing penetration of renewable energy brings unprecedented operational challenges to the modern power system, demanding additional power system flexibility to enhance grid reliability. This NSF CAREER project aims to bolster the nation’s grid decarbonization and energy security by developing a grid-edge resource (GER) management framework to increase power system flexibility. The project will bring transformative changes to GER management by overcoming three intertwined challenges, i.e., high scalability requirements, major privacy concerns, and surging cybersecurity risks. This will be achieved by synthesizing ideas from optimization, machine learning, statistics, and cryptology to establish an efficient, private, and secure GER control framework. The intellectual merits of the project include (1) developing a scalable framework that enables efficient cooperative control of heterogeneous GERs; (2) investigating privacy preservation measures for strongly coupled decentralized GER control; (3) investigating models, and detection and mitigation strategies of stealthy cyber-attacks that target at decentralized GER control algorithms. The broader impacts of the project include (1) unleashing heterogeneous GERs to enhance grid reliability and deepen grid decarbonization; (2) advancing the grid’s ability to integrate increasing amounts of renewable generation and GERs in a cost-effective, secure, and reliable way; (3) providing the industry with insight into developing new market products. The integrated education plan will spread control and power system concepts to youth-in-custody, K-12 students, and underrepresented groups, motivating them to pursue STEM education and careers. Existing GER control frameworks may fail in large-scale deployment because they invariably ignore scalability issues caused by network dimension and GER heterogeneity, lack efficient measures to protect GER owners’ privacy, and lack the understanding of stealthy for-purpose cyber-attacks. To advance the knowledge, this project will (1) develop new optimization, machine learning, and statistical tools to construct a GER control framework that is agnostic to GER type, scalable with respect to GER population and network dimension, and applicable for different grid services; (2) characterize privacy risks in GER management and develop cryptology-based and non-cryptology-based privacy-preserving decentralized optimization paradigms; (3) constitute cyber-attack vectors that leverage decentralized GER control algorithms for stealthy attack purposes, determine the detectability of those attacks, and develop corresponding detection and mitigation strategies.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.