CDS&E: Extracting Models from Data - A Novel Data-Driven Simulation Strategy for Reacting Flows

CDS

• 批准号: 1953350
• 负责人: James Sutherland
• 金额: $ 45.32万
• 依托单位: University of Utah
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1953350&HistoricalAwards=false
• 关键词: CDS& Extracting; Models; Data; Novel

Numerical simulation of reacting flows of real systems is computationally challenging because of the large number variables involved. This research will develop a new data-driven modelling approach that directly incorporates information from canonical or reference test cases to extract simplified models with user-defined error limits. In particular, novel machine learning concepts will be used to generate simplified models that are suitable for use in practical, engineering-scale simulations. Such computationally efficient models can have a direct impact in addressing a range of relevant energy and environmental problems, for example oxy-fuel combustion (for easier CO2 capture and sequestration), pollutant and particulate formation in stationary and mobile combustion systems, etc. The techniques developed here are also applicable to other fields where many reaction manifolds or pathways exist such as in plasma physics or atmospheric chemistry.Many systems in nature evolve along manifolds, which are smooth and reduced complexity subspaces of a parameter space which satisfy, often unknown, physical constraints. However, developing models to describe this evolution is challenging. A key challenge to this modeling approach is dealing with the source terms that arise in the reduced order model. These are a reflection of the source terms in the full (high-fidelity) model, but must be well-parameterized by the reduced-order model parameters without causing unphysical behavior like divergence near manifold boundaries and spurious source/sink points. This will harness the power of data science to characterize the geometry of the low-dimensional manifold and use that information to improve the behavior of the derived models. Extracting the low-dimensional model is challenging because it requires identifying a moderate dimensional shape where gridding and meshing techniques which scale exponentially in dimension will fail. This work will instead be limited by properties such as boundary curvature and vector field acceleration which are well-controlled for most physics-defined systems. Moreover, these models will be learned in a robust manner consistent with a simple physical model, in spite of noise in training data which may otherwise result in spurious critical saddle points in the resulting vector field.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.

由于涉及大量变量，真实系统反应流的数值模拟在计算上具有挑战性。这项研究将开发一种新的数据驱动建模方法，该方法直接结合来自规范或参考测试用例的信息，以提取具有用户定义的错误限制的简化模型。特别是，新颖的机器学习概念将用于生成适用于实际工程规模模拟的简化模型。这种计算效率高的模型可以对解决一系列相关的能源和环境问题产生直接影响，例如富氧燃料燃烧（更容易捕获和封存二氧化碳）、固定和移动燃烧系统中污染物和颗粒的形成等。这里开发的也适用于存在许多反应流形或路径的其他领域，例如等离子体物理或大气化学。自然界中的许多系统沿着流形演化，这些流形是参数空间的平滑且复杂性降低的子空间，满足通常未知的物理限制。 然而，开发模型来描述这种演变具有挑战性。这种建模方法的一个关键挑战是处理降阶模型中出现的源项。这些是完整（高保真）模型中源项的反映，但必须通过降阶模型参数进行良好参数化，而不会导致流形边界附近的发散和虚假源/汇点等非物理行为。这将利用数据科学的力量来表征低维流形的几何形状，并使用该信息来改进派生模型的行为。提取低维模型具有挑战性，因为它需要识别中等维度的形状，而在该形状中，尺寸呈指数级扩展的网格和网格划分技术将失败。相反，这项工作将受到边界曲率和矢量场加速度等属性的限制，这些属性对于大多数物理定义的系统来说都可以得到很好的控制。此外，这些模型将以与简单物理模型一致的鲁棒方式进行学习，尽管训练数据中存在噪声，否则可能会导致结果向量场中出现虚假的临界鞍点。该奖项反映了 NSF 的法定使命，并被视为值得通过使用基金会的智力优点和更广泛的影响审查标准进行评估来支持。

项目成果

Manifold-informed state vector subset for reduced-order modeling

用于降阶建模的流形通知状态向量子集

• DOI: 10.1016/j.proci.2022.06.019
• 发表时间: 2022
• 期刊: Proceedings of the Combustion Institute
• 影响因子: 3.4
• 作者: Zdybał, Kamila;Sutherland, James C.;Parente, Alessandro
• 通讯作者: Parente, Alessandro

Batch Multi-Fidelity Active Learning with Budget Constraints

具有预算约束的批量多保真主动学习

• 发表时间: 2022
• 期刊: Advances in neural information processing systems
• 作者: Shibo Li, Jeff Phillips

Local manifold learning and its link to domain-based physics knowledge

局部流形学习及其与基于领域的物理知识的联系

• DOI: 10.1016/j.jaecs.2023.100131
• 发表时间: 2023
• 期刊: Applications in Energy and Combustion Science
• 作者: Zdybał, Kamila;D’Alessio, Giuseppe;Attili, Antonio;Coussement, Axel;Sutherland, James C.;Parente, Alessandro
• 通讯作者: Parente, Alessandro

A technique for characterising feature size and quality of manifolds

• DOI: 10.1080/13647830.2021.1931715
• 发表时间: 2021-06-03
• 期刊: COMBUSTION THEORY AND MODELLING
• 影响因子: 1.3
• 作者: Armstrong, Elizabeth;Sutherland, James C.
• 通讯作者: Sutherland, James C.

PCAfold 2.0—Novel tools and algorithms for low-dimensional manifold assessment and optimization

PCAfold 2.0——用于低维流形评估和优化的新颖工具和算法

• DOI: 10.1016/j.softx.2023.101447
• 发表时间: 2023
• 期刊: SoftwareX
• 影响因子: 3.4
• 作者: Zdybał, Kamila;Armstrong, Elizabeth;Parente, Alessandro;Sutherland, James C.
• 通讯作者: Sutherland, James C.