CAREER: Modeling and Quantification of the Interdependent Power Grid Uncertainties

NSF CAREER项目旨在解决日益增加的电网不确定性带来的挑战。间歇性可持续能源的预期增长将给电网带来不可避免的不确定性。与此同时，近年来恶劣的天气模式显著增加了电网的中断和组件故障。迫切需要对这些高度相互依赖的不确定性进行建模和量化的新技术。这些技术将使未来的电网更好地为即将到来的不确定性激增做好准备，并提高电网的可靠性。该项目将为实时工具带来革命性的变化，这些工具可以对相互依赖的电网不确定性（即可再生能源发电和停电）的影响进行建模和量化。该项目的智力优势包括引入电网不确定性建模和量化及其对电网运行的因果影响的新知识。该项目的更广泛影响包括在不确定条件下提高电网运行的效率和可靠性，解决可持续能源整合的障碍，以及应对气候变化的挑战。该项目的综合研究和教育目标将培养来自不同群体的学生，包括本科生和高中生。通过与行业专家、公共图书馆和当地学校的合作，教育和推广活动促进了工程领域中代表性不足的群体的参与和保留。该项目将开发有效的自适应技术，在考虑到相互依赖的网格不确定性的快速演变性质的同时，对其进行建模和量化。由于分析电力系统非线性动力学和大规模历史数据的复杂性，特别是在近实时情况下，现有的电网不确定性量化工作受到限制。该项目将融合级联故障建模、机器学习和统计等工具，开发一个可扩展且高效的量化框架，以了解发电和停电不确定性之间的相互依赖关系，并允许进行近实时分析。为了更现实地量化电网的不确定性，数据驱动模型和物理潮流方程将相互通知以产生混合随机模型。在这些模型的基础上，将开发新的因果模型来演示个体不确定性如何影响电网运行。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。

This NSF CAREER project aims to address the challenges introduced by the increasing power grid uncertainties. The expected increase in intermittent sustainable energy resources will introduce inevitable uncertainties to the power grids. At the same time, the severe weather patterns in recent years have significantly increased the grid outages and component failures. There is a critical need for new techniques that model and quantify these highly interdependent uncertainties. Such techniques will better prepare future grids for the impending surge in uncertainties and facilitate grid reliability. This project will bring transformative changes to real-time tools that model and quantify the impacts of interdependent grid uncertainties, namely, renewable generation and outages. The intellectual merits of the project include the introduction of new knowledge on modeling and quantification of power grid uncertainties and their causal impacts on grid operations. The broader impacts of the project include more efficient and reliable grid operations under uncertainties, tackling barriers for the integration of sustainable energy resources, and addressing the challenges of climate change. The integrated research and education objectives of the project will train students from diverse groups, including undergraduate and high school students. The education and outreach activities promote the participation and retention of the underrepresented groups in engineering through collaborations with industry experts, public libraries, and local schools.The project will develop efficient and adaptive techniques to model and quantify interdependent grid uncertainties while considering their fast-evolving nature. The existing efforts on grid uncertainty quantification have been limited due to the complexity of analyzing the nonlinear dynamics of power systems and large-scale historical data, particularly in near real-time. This project will fuse tools from cascading failure modeling, machine learning, and statistics to develop a scalable and efficient quantification framework to understand the interdependencies among generation and outage uncertainties and allow for near real-time analysis. For a more realistic quantification of the grid uncertainties, the data-driven models and physical power flow equations will inform each other to yield hybrid stochastic models. Building upon these models, new causal models will be developed to demonstrate how individual uncertainties impact grid operations.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.