AMPS: Advanced Mathematical Algorithms for Model Reduction and Stochastic Modeling for the Emerging Power Grid

AMPS：用于新兴电网模型简化和随机建模的高级数学算法

• 批准号: 1734727
• 负责人: Barry Lee
• 金额: $ 24万
• 依托单位: Southern Methodist University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-09-01 至 2023-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1734727&HistoricalAwards=false
• 关键词: AMPS; Advanced; Mathematical; Algorithms; Model

This project will develop tools to analyze and help design robust and resilient modern power grid systems. The emergent power grid is very different from the traditional grid, which is based on technology dating back to the early 20'th century. With new technologies and regulatory policies, new challenges are arising in the design of stable power systems that can reliably deliver electric power. To tackle these challenges, mathematical and computational tools are needed to analyze these power grids with the new features. The goal of this project is to develop these mathematical and computational tools. Specifically, this project will examine model reduction, uncertainty/stochasticity, and stability, which are some of the new features that must be incorporated in the models of the modern grid. Stability will be the connecting theme in these topics since stability is required to ensure a resilient grid with reduced risks. The PI will examine (1) how synchrony in the oscillations of the grid can be used to determine coherent sets of generators and loads for accurate model reduction; (2) what are the relevant structures that must be reserved to produce accurate reduced models; (3) how to develop multilevel solvers for decentralized systems; (4) how stochasticity affects the synchrony, coherency, model reduction, and stability of the grid; and (5) how observed data can be assimilated into the stochastic models. The stochasticity will be incorporated using stochastic differential-algebraic equations. This stochasticity will be introduced using the Orstein-Uhlenbeck process that best describes the stochasticity in the components of the system.

该项目将开发分析和帮助设计强大而有弹性的现代电网系统的工具。新兴的电网与传统电网大不相同，该电网基于20世纪初的技术。借助新技术和法规政策，在可以可靠地提供电力的稳定电力系统的设计中产生了新的挑战。为了应对这些挑战，需要数学和计算工具来通过新功能分析这些电网。该项目的目的是开发这些数学和计算工具。 具体而言，该项目将检查降低模型，不确定性/随机性和稳定性，这些新功能必须纳入现代网格模型中。稳定性将是这些主题中的连接主题，因为需要稳定性以确保风险降低的弹性网格。 PI将检查（1）如何使用网格振荡中的同步来确定发电机和载荷的相干集，以进行准确的模型还原； （2）必须保留哪些相关结构才能产生准确的减少模型； （3）如何为分散系统开发多级求解器； （4）随机性如何影响网格的同步，相干性，模型还原和稳定性； （5）如何将观察到的数据吸收到随机模型中。随机差异代数方程式将纳入随机性。这种随机性将使用Orstein-Uhlenbeck过程进行引入，该过程最能描述系统组件中的随机性。

项目成果

Multigrid for model reduction of power grid networks

• DOI: 10.1002/nla.2201
• 发表时间: 2018-07
• 期刊: Numerical Linear Algebra with Applications
• 影响因子: 4.3
• 作者: Barry Lee

Bringing physics into the coarse‐grid selection: Approximate diffusion distance/effective resistance measures for network analysis and algebraic multigrid for graph Laplacians and systems of elliptic partial differential equations

将物理学引入粗网格选择：用于网络分析的近似扩散距离/有效阻力测量以及用于图拉普拉斯算子和椭圆偏微分方程组的代数多重网格

• DOI: 10.1002/nla.2539
• 发表时间: 2023
• 期刊: Numerical Linear Algebra with Applications
• 影响因子: 4.3
• 作者: Lee, Barry

Algebraic multigrid for the nonlinear powerflow equations

• DOI: 10.1002/nla.2347
• 发表时间: 2020-11
• 期刊: Numerical Linear Algebra with Applications
• 影响因子: 4.3
• 作者: Barry Lee;Enrique Pereira Batista

Algebraic multigrid for systems of elliptic boundary‐value problems

椭圆边界值问题系统的代数多重网格

• DOI: 10.1002/nla.2303
• 发表时间: 2021
• 期刊: Numerical Linear Algebra with Applications
• 影响因子: 4.3
• 作者: Lee, Barry