CAREER: Estimation and Control of Electrochemical-Thermal Battery Models: Theory and Experiments

职业：电化学热电池模型的估计和控制：理论和实验

• 批准号: 1847177
• 负责人: Scott Moura
• 金额: $ 50万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-03-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1847177&HistoricalAwards=false
• 关键词: CAREER; Estimation; Control; Electrochemical; Thermal

This Faculty Early Career Development Program (CAREER) project will benefit national interests by advancing knowledge on battery management systems based on electrochemical-thermal models. Batteries are the linchpin technology for multiple economic sectors, including consumer electronics, transportation, and electric power systems. However, today's battery management systems use simplistic models, which have raised serious performance and safety issues. For example, significant electrification of the U.S. vehicle fleet will require fast charging and long-range batteries. Simultaneously, we must ensure safety, as evidenced by recent cases where batteries have caught fire. Future battery management systems will address these deficiencies and unlock increased performance and safety by utilizing high-fidelity multi-physics models. However, the electrochemical-thermal model dynamics present unsolved challenges for estimation and control. The research goal of this project is to resolve these challenges and generate results that will enable current and future batteries with more energy, more power, faster charge times, and longer life. The educational goal of this project is to enhance retention and performance among students from underrepresented, low-income, and first-generation backgrounds. This will be achieved through a "Maker Design Studio," which will train over 600 Science, Technology, Engineering, and Mathematics (STEM) students to become the next generation of energy and control engineering leaders.Batteries are characterized by multi-physics mathematical models, often involving nonlinear Partial Differential Equations (PDEs), limited sensing and actuation, and significant parameter uncertainty. This project pursues three research goals, motivated by batteries yet in pursuit of fundamental systems and control challenges: (1) Formulate and analyze a parameter estimation framework, based on a data selection approach that resolves the identifiability problem. Online battery parameter (i.e. state-of-health) estimation has remained elusive, due to fundamental identifiability challenges. (2) Experimentally quantify the benefits of an electrochemical model-based battery management system in terms of fast charge times and capacity loss. Today, it is unclear if rigorously designed electrochemical model-based management systems yield significant improvements, due to the lack of experimental evidence. This project leverages a unique battery-in-the-loop testbed to reveal the true impact of electrochemical-based management methods. (3) Create a PDE-based analysis, estimation, and control framework for coupled parabolic-hyperbolic PDEs, with application to battery thermal management. Specifically, the project pursues a weak-variations approach to design linear quadratic estimators and controllers. Overall, this project focuses on fundamental advancements to estimation and control that will accelerate a paradigm shift toward multi-physics control-theoretic battery management systems that will enable a new generation of energy storage.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.

该教师早期职业发展计划（CAREER）项目将通过推进基于电化学-热模型的电池管理系统知识，使国家利益受益。电池是多个经济部门的关键技术，包括消费电子产品，运输和电力系统。然而，今天的电池管理系统使用简单的模型，这已经引起了严重的性能和安全问题。例如，美国车队的电气化将需要快速充电和长距离电池。与此同时，我们必须确保安全，最近发生的电池起火事件就是证明。未来的电池管理系统将解决这些缺陷，并通过利用高保真多物理模型来提高性能和安全性。然而，电化学-热模型动力学的估计和控制提出了未解决的挑战。该项目的研究目标是解决这些挑战，并产生结果，使当前和未来的电池具有更多的能量，更多的功率，更快的充电时间和更长的寿命。该项目的教育目标是提高来自代表性不足，低收入和第一代背景的学生的保留率和表现。这将通过“创客设计工作室”实现，该工作室将培养600多名科学、技术、工程和数学（STEM）学生，使其成为下一代能源和控制工程的领导者。电池的特点是多物理数学模型，通常涉及非线性偏微分方程（PDE）、有限的传感和驱动以及显著的参数不确定性。该项目追求三个研究目标，由电池驱动，但在追求基本系统和控制的挑战：（1）制定和分析参数估计框架，基于解决可识别性问题的数据选择方法。由于基本的可识别性挑战，在线电池参数（即健康状态）估计仍然难以捉摸。(2)通过实验量化基于电化学模型的电池管理系统在快速充电时间和容量损失方面的优势。今天，由于缺乏实验证据，尚不清楚严格设计的基于电化学模型的管理系统是否会产生显着的改善。该项目利用独特的电池在环测试平台来揭示基于电化学的管理方法的真正影响。(3)为耦合抛物-双曲偏微分方程创建基于偏微分方程的分析、估计和控制框架，并将其应用于电池热管理。具体而言，该项目追求弱变化的方法来设计线性二次估计器和控制器。总体而言，该项目侧重于评估和控制方面的根本性进展，这将加速向多物理场控制理论电池管理系统的范式转变，从而实现新一代储能。该奖项反映了NSF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。

项目成果

Global Sensitivity Analysis of 0-D Lithium Sulfur Electrochemical Model

0-D锂硫电化学模型的全局灵敏度分析

• 发表时间: 2023
• 期刊: IFAC World Congress 2023
• 作者: Dangwal, C;Kato, D.;Huang, Z.;Kandel, A.;Moura, S. J.
• 通讯作者: Moura, S. J.

Estimation of Parameter Probability Distributions for Lithium-Ion Battery String Models Using Bayesian Methods

• DOI: 10.1115/dscc2020-3218
• 发表时间: 2020-10
• 期刊: Volume 1: Adaptive/Intelligent Sys. Control; Driver Assistance/Autonomous Tech.; Control Design Methods; Nonlinear Control; Robotics; Assistive/Rehabilitation Devices; Biomedical/Neural Systems; Building Energy Systems; Connected Vehicle Systems; Control/Estimation of Energy Systems; Control Apps.;
• 作者: Luis D. Couto;Dong Zhang;A. Aitio;S. Moura;D. Howey

State Estimation for a Zero-Dimensional Electrochemical Model of Lithium-Sulfur Batteries

• DOI: 10.23919/acc50511.2021.9483225
• 发表时间: 2021-01
• 期刊: 2021 American Control Conference (ACC)
• 作者: Zhijia Huang;Dong Zhang;Luis D. Couto;Quan-hong Yang;S. Moura

Reinforcement Learning-based Fast Charging Control Strategy for Li-ion Batteries

• DOI: 10.1109/ccta41146.2020.9206314
• 发表时间: 2020-02
• 期刊: 2020 IEEE Conference on Control Technology and Applications (CCTA)
• 作者: Saehong Park;Andrea Pozzi;Michael Whitmeyer;Won Tae Joe;D. Raimondo;S. Moura

Integrating Electrochemical Modeling with Machine Learning for Lithium-Ion Batteries

• DOI: 10.23919/acc50511.2021.9482997
• 发表时间: 2021-03
• 期刊: 2021 American Control Conference (ACC)
• 作者: H. Tu;S. Moura;H. Fang