CAREER: Design of Cellular Mechanical Metamaterials under Uncertainty with Physics-Informed and Data-Driven Machine Learning

该职业项目的目标是吸引和教育研究生和本科生进行材料设计，特别关注在制造相关材料不确定性的影响下设计细胞机械超材料（CMM）。该计划包括基于物理的计算、机器学习（ML）和不确定性策略下的设计，以及外展活动的发展。基本假设是，可以通过开发多尺度计算框架来研究坐标测量机的设计，以实现目标机械特性和性能，该框架可研究组件尺度特性与底层微尺度架构之间的关系。该项目的社会影响将对经济产生影响，并有望设计出可持续、轻质和高性能的材料。所获得的知识将通过技术活动和开放获取的图形软件工具传播到学术界和工业界。该项目的其他可交付成果包括本科生和研究生级别的课程开发、学生的研究经验以及涉及学生和教育工作者的其他外展活动，特别关注来自代表性不足群体的个人。该项目的总体目标是使用多尺度框架提高 CMM 设计的现有知识并提高 3D 打印产品的性能，该框架将通过允许非周期性重复的微结构设计并考虑制造相关的不确定性来探索微结构和组件之间的复杂和非线性关系。这一目标将通过开发由基于物理的材料模型、数据驱动和物理信息的机器学习、设计优化和不确定性量化方法驱动的多尺度设计策略来实现。对坐标测量机非周期性微观结构排列进行建模的能力对于探索其真正的部件级机械性能至关重要，从而大大增加其在高超音速、结构应用、能量吸收、传感器和软机器人等新一代工程系统中的使用潜力。该项目的研究结果还将确定通过考虑材料不确定性的影响来提高机械性能和可靠性的设计。此外，坐标测量机的设计方法将扩展到受自然启发的细胞材料，例如人造骨结构，用于设计此类系统以在不确定性下实现目标机械性能。该活动还将通过开展外展活动来促进教学、培训和学习，例如针对青少年和教师的夏令营、项目和讲习班。通过专门针对女学生、第一代大学生、来自弗吉尼亚州西南部服务不足社区的学生和其他少数群体的外展计划，保证了代表性不足的群体的参与。该项目数据和研究结果将在弗吉尼亚理工大学的开放获取存储库 VTechData 上公开提供。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

The objective of this CAREER project is to engage and educate graduate and undergraduate students in materials design, with a particular focus on designing cellular mechanical metamaterials (CMMs) under the effects of fabrication-related material uncertainty. The plan includes physics-based computations, machine learning (ML), and design under uncertainty strategies, as well as the development of outreach activities. The underlying hypothesis is that the CMMs can be designed to achieve targeted mechanical properties and performance by developing a multi-scale computational framework that investigates the relationship between component-scale properties and underlying micro-scale architectures. The societal impacts of the project will be on the economy, with the promise of designing sustainable, lightweight, and high-performance materials. The gained knowledge will be disseminated to academia and industry through technical activities and open-access graphical software tools. Additional deliverables of the project include curriculum development at undergraduate and graduate levels, research experiences for students, and other outreach activities involving students and educators, with a special focus on individuals from underrepresented groups.The overarching goal of this project is to improve the current knowledge of CMM design and enhance the performance of 3-D printed products using a multi-scale framework that will explore complex and non-linear relationships between the microstructure and component by allowing non-periodically repeating microstructure designs and accounting for the fabrication-related uncertainty. This goal will be accomplished by developing a multi-scale design strategy driven by physics-based material models, data-driven and physics-informed ML, design optimization, and uncertainty quantification approaches. The ability to model non-periodical microstructure arrangements of CMMs will be essential to explore their true component-level mechanical performance, thereby substantially increasing their potential for use in new-generation engineering systems for hypersonics, structural applications, energy absorption, sensors, and soft robots. The findings of the project will also identify designs that improve mechanical performance and reliability by considering the effects of material uncertainty. In addition, the design methodology for CMMs will be extended to nature-inspired cellular materials, such as artificial bone structures, for designing such systems to achieve target mechanical performance under uncertainty. The activity will also promote teaching, training, and learning through the development of outreach activities, such as camps, programs, and workshops targeting both youths and teachers. The participation of underrepresented groups is guaranteed by specifically addressing outreach programs for female students, first-generation college students, students from underserved communities in Southwest Virginia, and other minorities. The project data and findings will be made publicly available at Virginia Tech’s open-access repository, VTechData.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.