Workshop: Applications of Machine Learning to Experimental Mechanics and Materials; Arlington, Virginia; 24-25 September 2019

研讨会：机器学习在实验力学和材料中的应用；

• 批准号: 1940102
• 负责人: Ioannis Chasiotis
• 金额: $ 5万
• 依托单位: University of Illinois at Urbana-Champaign
• 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: 2019-08-01 至 2020-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1940102&HistoricalAwards=false
• 关键词: Workshop; Applications; Machine; Learning; Experimental

This grant supports a workshop to explore the use of machine learning with experimental mechanics and materials. The rapid development of new materials with complex processing and structure, along with the emergence of new and potent computer-assisted experimental methods for materials and mechanical characterization have led to challenges in effectively and efficiently analyzing very large data sets in order to determine the important parameters that control the overall mechanical behavior. While established and detailed methodologies, grounded on materials physics and mechanics, serve as the foundation to evaluate and design new classes of materials with desirable mechanical properties, Data Science driven approaches for rapid assessment of important problem parameters could accelerate materials development and facilitate transitions through rapid processes for identifying structure-mechanical properties relationships from large experimental and modeling data sets. Such capabilities can impact new and emerging manufacturing methods, e.g. additive manufacturing, help us to understand complex processes in biological material systems, and finally accelerate the design of mechanically robust material systems. This workshop aims at connecting mechanicians with researchers in the Data Sciences field for a dialogue that could open new avenues in the field of mechanics of materials, with special emphasis on the application of machine learning to experimental mechanics. It will bring together researchers who engage in new, multimodal, experimental methods in mechanics, with early adopters of machine learning tools in the fields of materials and mechanics, and leaders from the machine learning community. The workshop aims at developing a long term perspective for the introduction of Data Science methods to the field of mechanics of materials. Therefore, a specific aim is to assess the potential and the limitations of machine learning techniques in providing guidance and enhanced capabilities to quantify the contribution of the, often numerous and coupled, parameters governing the mechanics of complex materials and systems.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.

这项拨款支持一个研讨会，以探索机器学习与实验力学和材料的使用。具有复杂加工和结构的新材料的快速发展，以及用于材料和力学表征的新型和强大的计算机辅助实验方法的出现，导致了有效和高效地分析非常大的数据集以确定控制整体力学行为的重要参数的挑战。虽然建立和详细的方法，以材料物理和力学为基础，作为评估和设计具有理想机械性能的新型材料的基础，但数据科学驱动的方法用于快速评估重要问题参数，可以加速材料的开发，并通过快速过程从大型实验和建模数据集中识别结构-机械性能关系，促进过渡。这种能力可以影响新的和新兴的制造方法，例如增材制造，帮助我们理解生物材料系统中的复杂过程，并最终加速机械坚固材料系统的设计。本次研讨会旨在将力学家与数据科学领域的研究人员联系起来，进行对话，以开辟材料力学领域的新途径，特别强调机器学习在实验力学中的应用。它将汇集从事新的，多模态，力学实验方法的研究人员，材料和力学领域机器学习工具的早期采用者，以及机器学习社区的领导者。本次研讨会旨在为材料力学领域的数据科学方法的引入发展一个长期的视角。因此，一个特定的目标是评估机器学习技术在提供指导和增强能力方面的潜力和局限性，以量化控制复杂材料和系统力学的众多和耦合参数的贡献。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。