ERI: Multilevel Inverse Robust Co-Design of Materials, Products, and Manufacturing Processes

本工程研究启动（ERI）项目的研究目标是研究在材料和产品设计的早期阶段共同考虑制造过程的新方法。通过在设计的早期阶段考虑制造过程，设计师可以做出有效的决策来定制材料微观结构并实现所需的产品性能，正如集成计算材料工程（ICME）所设想的那样。当前的材料设计方法侧重于在单个层面上依次探索基于模型的材料和产品关系，而不是在一个整体系统内同时实现产品、材料和制造。在这个项目中设想的整体的、基于系统的、强大的协同设计方法将考虑涉及产品和不确定制造过程的多层次材料系统的目标和相互作用。该方法将使设计师能够同时可视化和探索特定级别的目标、多级交互和子级别分析，从而实现数字设计、材料结构和制造工作流程的定量联系。该项目将通过战略性地整合系统工程、工程设计和材料科学领域的知识和技术来促进科学的进步，从而使人们对产品、材料和制造过程之间的耦合有一个基本的了解。此外，通过设计嵌入ICME设计主题的新课程，并组织相关工具的暑期学校，学生和从业者将从系统的角度更好地准备应对复杂的材料设计挑战。该ERI项目将建立将分层结构材料、产品及其制造过程与多目标系统设计策略和不确定性管理相耦合的多层次建模的基础技术。研究计划包括两个部分：(1)建立历史依赖的、数据驱动的处理-结构-属性-性能（PSPP）联系；(2)制定一个基于系统的逆协同设计框架，通过管理不确定性及其传播来探索跨多个层次的鲁棒解的范围集合。总体预期结果是一个新的基于系统的多层次框架，用于在不确定性下由分布式利益相关者做出的稳健决策方面对材料、产品和制造过程进行逆向协同设计。智力优势包括：(1)对多层材料、制造工艺和产品设计决策的相互依存性质有基本的理解；(2)实现产品的物质结构层次的有效数学表示；(3)在设计的早期阶段，为多级鲁棒协同设计可视化和探索不同工程学科（设计、材料和制造）之间的相互作用提供了一个计算框架。计划开展四项相关的教育和推广活动，将研究与更广泛的学生和专业人士群体结合起来：(1)通过整合现有研究生课程的结果，形式化数据驱动的PSPP链接建模程序；(2)通过新的ICME设计课程，迭代、更新和传播基于系统的标准ICME问题设计方法；(3)组织为期一周的ICME设计相关工具在线暑期学校；(4)通过科学出版物、报告和研究专著广泛传播研究成果。这些活动将导致新的经过验证和验证的技术来教授ICME设计；迭代、评估和建立构造的通用实用程序的新测试示例（单独的和集成的）；以及将多层材料建模，制造考虑因素和基于决策的设计策略注入ICME工作流程的基础。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。

The research objective of this Engineering Research Initiation (ERI) project is to investigate new ways to co-consider manufacturing processes in the early stages of materials and product design. By including the considerations of manufacturing processes in the early stages of design, designers can make effective decisions to customize material microstructures and achieve desired product performance, as envisioned by Integrated Computational Materials Engineering (ICME). Current materials design methods focus on sequentially exploring model-based relations of materials and products at individual levels instead of a concurrent realization of products, materials, and manufacturing within a holistic system. The holistic, systems-based, robust co-design approach envisioned in this project will account for the goals of and interactions between multilevel material systems involving the product and the manufacturing processes under uncertainty. The approach will enable designers to concurrently visualize and explore level-specific objectives, multilevel interactions, and sublevel analyses, thus enabling quantitative linkages in digital design to materials structure to manufacturing workflows. The project will promote the progress of science by strategically integrating knowledge and techniques from the fields of systems engineering, engineering design, and materials science to enable a fundamental understanding of the couplings between products, materials, and manufacturing processes. Additionally, by designing new curricula that embed Design for ICME topics and organizing a summer school on the associated tools, students and practitioners will be better prepared to address complex materials design challenges from a systems perspective.This ERI project will establish foundational techniques for coupling multilevel modeling of hierarchically structured materials, products, and their manufacturing processes with multiobjective systems design strategies and uncertainty management. The research plan consists of two parts: (1) establish history-dependent, data-driven processing-structure-property-performance (PSPP) linkages; and (2) formulate a systems-based inverse co-design framework for exploring ranged sets of robust solutions across multiple levels by managing uncertainty and its propagation. The overall expected outcome is a novel systems-based multilevel framework for the inverse co-design of materials, products, and manufacturing processes in terms of robust decisions made by distributed stakeholders under uncertainty. The intellectual merit includes: (1) a fundamental understanding of the interdependent nature of multilevel materials, manufacturing processes, and product design decisions; (2) an efficient mathematical representation of material structure hierarchy to realize products; and (3) a computational framework for the multilevel robust co-design visualization and exploration of the interactions between the different engineering disciplines (design, material, and manufacturing) in the early stages of design. Four related educational and outreach activities are planned to integrate the research with the broader community of students and professionals: (1) formalize the data-driven PSPP linkage modeling procedure by integrating the results in existing graduate courses; (2) iterate, update, and disseminate a systems-based design approach for standard ICME problems through a new Design for ICME course; (3) organize a one-week online summer school on tools relevant to Design for ICME; and (4) broadly disseminate results through scientific publications, presentations, and research monographs. These activities will lead to new verified and validated techniques to teach Design for ICME; new test examples to iterate, assess, and establish the generic utility of the constructs (individually and integrated); and a foundation for infusing multilevel materials modeling, manufacturing considerations, and decision-based design strategies into ICME workflows.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.