NRT-AI: Harnessing AI for Inverse Design Training in Advanced and Sustainable Composites (IDeAS Composites)

尽管复合材料有广阔的设计空间，但在性能、经济和环境目标与当前的设计和制造方法之间存在着巨大的差距。最令人震惊的是昂贵、漫长的开发周期，以及浪费资源并可能对环境和气候变化产生不利影响的次优设计。这些差距的根本原因是缺乏对材料结构、工艺方法和参数对材料微观结构演变的影响以及随后最终产品的物理、经济和环境性能的详细了解。这项国家科学基金研究培训(NRT)利用人工智能(AI)进行先进和可持续复合材料(IDEA)的逆向设计培训，将通过一个基于物理的、基于AI的建模和设计平台培训学生，该平台将使学生能够发现新的复合材料形式和相关的新制造方法。授予克莱姆森大学的NRT奖将通过由学术教师和行业研究人员共同指导的变革性AI时代课程，催化学员研究和发现途径的转变。Ideas Composites计划将培训总共50名学生；其中25人将是NRT资助的IDEAS研究员，其余25人将被确定为IDEAS学者。该项目将吸引来自计算机科学、数据科学、统计科学、机械工程、汽车工程和材料科学的学员，并将赋予学员学术界和行业联合培训的技能集，以确保他们在人工智能时代取得成功。该NRT项目的研究主题集中在发现和调查基于物理知识、基于机器学习的逆向设计平台在开发新的复合材料体系结构和制造方法方面的有效性。该计划将培养一批研究生，他们拥有深厚的专业知识，并得到广泛的交叉技能知识的支持，并在学术和行业专家的协作下，为他们配备独特的“DNA形状”技能集。具体地说，该计划将(1)通过构建一套模拟复合材料部件生命周期的高保真模型--“数字生命周期”，研究机器学习方法在复合材料逆结构和制造过程设计中的应用，并开发材料和制造集成设计的逆向设计方法，在人工智能与复合材料逆向设计和制造创新的交叉点促进跨学科研究；(2)探索研究生和本科生的联合培养模式，包括复合材料逆向设计顶峰和研究设计、开发和示范(RD&D；D)项目，以应用于行业问题的研究成果为中心；(3)创造一个多样化、公平和包容的环境，促进跨学科合作，使受训者为需要独特的“DNA形状”技能的职业做好准备；以及(4)建立跨学科教育计划，以(A)培养下一代复合材料工程毕业生(到第5年总共50人)，他们将拥有必要的人工智能逆向设计专业知识和技能，以应对即将到来的人工智能时代的独特挑战，并在复合材料行业蓬勃发展，以及(B)培训现有的劳动力，以增强他们的知识，并促进人工智能方法和原理在复合材料工程界的传播。NSF研究培训(NRT)计划旨在鼓励开发和实施大胆的、具有潜在变革意义的STEM研究生教育培训模式。该计划致力于通过创新的、基于证据的、与不断变化的劳动力和研究需求保持一致的综合培训模式，在高度优先的跨学科或趋同的研究领域有效地培训STEM研究生。  该奖项反映了国家科学基金会的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Despite the vast design space of composites, there are significant gaps between the performance, economic, and environmental targets and current design and manufacturing approaches. Most egregious are the expensive, long development cycles and the sub-optimal design that waste resources and may adversely affect the environment and climate change. The fundamental cause of such gaps is the lack of detailed understanding of the influence of the material architecture, process methods, and parameters on material microstructure evolution and subsequently the end product’s physical, economic, and environmental performance. This National Science Foundation Research Traineeship (NRT), harnessing artificial intelligence (AI) for Inverse Design Training in Advanced and Sustainable Composites (IDeAS), will train students through a physics-informed, AI-based modeling and design platform which will enable the discovery of new composites materials forms and relevant new manufacturing methodologies. This NRT award to Clemson University will catalyze a shift in the research and discovery pathway of the trainees via a transformative AI-age curriculum co-instructed by academic faculty and industry researchers. The IDeAS Composites program will train a total of 50 students; of these, 25 will be NRT-funded IDeAS fellows and the remaining 25 would be identified as IDeAS scholars. The program will draw trainees from computer science, data science, statistical science, mechanical engineering, automotive engineering, and materials science and will empower trainees with an academia–industry co-trained skill set that will ensure their success in the AI age. The research theme of this NRT program is focused on discovering and investigating the effectiveness of a physics-informed, machine-learning-based inverse design platform for developing new composite material architectures and manufacturing methodologies. The program will train a cohort of graduate students with deep, specialized expertise supported by broad, cross-skill knowledge, and equip them with a unique “DNA-shaped” skill set collaboratively facilitated by both academic and industry experts. Specifically, the program will (1) catalyze interdisciplinary research at the intersection of AI and the inverse design of composites and manufacturing innovation via constructing a “digital life cycle” which is a suite of high-fidelity models for simulating a composite component’s life cycle, investigating the application of machine-learning methods for inverse composite material architecture and manufacturing process design, and developing an inverse design approach for integrated material and manufacturing design; (2) explore a combined graduate and undergraduate student training model comprising a composites inverse design capstone and a research design, development, and demonstration (RD&D) project centering on research outcomes applied to industry problems; (3) create a diverse, equitable, and inclusive environment fostering interdisciplinary collaboration in which trainees are prepared for careers requiring a unique “DNA-shaped” skill set; and (4) establish an interdisciplinary education program to (a) prepare next-generation composites engineering graduates (altogether 50 by year 5) who will have AI-enabled inverse design expertise and skills necessary to meet the unique challenges of the coming AI age and to thrive in the composites industry, and (b) train the current workforce to enhance their knowledge and foster dissemination of AI methods and principles in the composites engineering community.The NSF Research Traineeship (NRT) Program is designed to encourage the development and implementation of bold, new potentially transformative models for STEM graduate education training. The program is dedicated to effective training of STEM graduate students in high priority interdisciplinary or convergent research areas through comprehensive traineeship models that are innovative, evidence-based, and aligned with changing workforce and research needs.  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.