FMRG: Manufacturing USA: Cyber: Data-Driven Methods for Future Cyber Manufacturing as a Service

FMRG：美国制造：网络：未来网络制造即服务的数据驱动方法

• 批准号: 2229260
• 负责人: Shreyes Melkote
• 金额: $ 300万
• 依托单位: Georgia Tech Research Corporation
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-15 至 2026-08-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2229260&HistoricalAwards=false
• 关键词: FMRG; Manufacturing; USA; Cyber; Data

Cyber manufacturing anticipates a networked service for on-demand manufacturing of engineered parts and products. Such a service has the potential to democratize access to the means and know-how of discrete parts manufacturing at the reduced time, scale, and cost. Advances in digital design, low-cost sensing, and networked manufacturing machines have potential to enable ready access to the means of production, along with design and production data. However, computational methods and tools that enable a future cyber manufacturing service to automatically translate a digital design into a manufactured product using the available means of production are needed. This award supports fundamental research to automate the selection, planning and acquisition of the manufacturing processes and resources required to produce discrete parts and products on-demand. This capability is applicable to a wide range of manufacturing industries including aerospace, automotive, energy, and biomedical. Therefore, this research will benefit the competitiveness of the U.S. manufacturing sector and society. The research integrates several disciplines including design, manufacturing science and technology, human-computer interaction, artificial intelligence, and machine learning. Through partnership with a Hispanic Serving Institution, industry, a Manufacturing USA institute, and outreach to other underrepresented communities, the project will educate and train a diverse future cyber manufacturing workforce. This project will research a data-driven computational framework that enables computers to infer design elements and processing details for new parts from the accumulated design and production experience of industry with parts that have been manufactured previously. This capability rests on extracting core manufacturing intelligence in the form of scalable and generative manufacturing process capability knowledge and design-for-manufacturing knowledge from past and current design and manufacturing data. Specifically, the research will yield fundamental knowledge of the generative capability of manufacturing processes to transform part shape, material properties, and quality to meet product design requirements, and generative redesign-for-functionality and manufacturability. Process planning and part redesign methods will be developed from a combination of new deep learning, shape descriptor, and manufacturing reasoning technologies. The resulting data-driven computational framework will provide the fundamental software tools to implement future cyber manufacturing services that provide a scalable and efficient solution for automatically translating digital designs into physical products on-demand.This Future Manufacturing award was supported by the Office of the Assistant Director (OAD) and the Division of Civil, Mechanical and Manufacturing Innovation (CMMI) of the Directorate for Engineering (ENG), the Division of Undergraduate Education (DUE) of the Directorate for Education and Human Resources (EHR) and the Division of Computer and Network Systems (CNS) of the Directorate for Computer and Information Science and Engineering (CISE).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.

网络制造为工程零件和产品的按需制造提供网络化服务。这种服务有可能在减少时间、规模和成本的情况下，使离散部件制造的手段和技术大众化。数字设计、低成本传感和网络化制造机器的进步有可能使生产资料以及设计和生产数据随时可用。然而，需要计算方法和工具，使未来的网络制造服务能够使用可用的生产手段自动将数字设计转化为制造产品。该奖项支持自动化选择、计划和获取制造过程和资源的基础研究，以按需生产离散零件和产品。这种能力适用于广泛的制造行业，包括航空航天、汽车、能源和生物医学。因此，这项研究将有利于美国制造业和社会的竞争力。该研究整合了多个学科，包括设计、制造科学与技术、人机交互、人工智能和机器学习。通过与拉美裔服务机构、行业、美国制造业机构的合作，以及与其他代表性不足的社区的合作，该项目将教育和培训多元化的未来网络制造业劳动力。该项目将研究一个数据驱动的计算框架，使计算机能够从积累的工业设计和生产经验中推断新零件的设计元素和加工细节。这种能力依赖于从过去和当前的设计和制造数据中以可扩展和生成的制造过程能力知识和面向制造的设计知识的形式提取核心制造智能。具体而言，该研究将产生制造过程生成能力的基础知识，以改变零件形状，材料属性和质量以满足产品设计要求，以及生成功能和可制造性的再设计。工艺规划和零件重新设计方法将从新的深度学习、形状描述符和制造推理技术的组合中发展出来。由此产生的数据驱动计算框架将提供基本的软件工具，以实现未来的网络制造服务，为按需自动将数字设计转换为物理产品提供可扩展和高效的解决方案。这项未来制造奖得到了助理主任办公室（OAD）和工程局（ENG）土木、机械和制造创新司（CMMI）、教育和人力资源局（EHR）本科教育司（DUE）以及计算机和信息科学与工程局（CISE）计算机和网络系统司（CNS）的支持。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。

项目成果

Manufacturing process selection based on similarity search: incorporating non-shape information in shape descriptor comparison

基于相似性搜索的制造工艺选择：在形状描述符比较中纳入非形状信息

• DOI: 10.1007/s10845-024-02368-5
• 发表时间: 2024
• 期刊: Journal of Intelligent Manufacturing
• 影响因子: 8.3
• 作者: Wang, Zhichao;Yan, Xiaoliang;Bjorni, Jacob;Dinar, Mahmoud;Melkote, Shreyes;Rosen, David
• 通讯作者: Rosen, David

Generative Design by Embedding Topology Optimization into Conditional Generative Adversarial Network

将拓扑优化嵌入条件生成对抗网络的生成设计

• DOI: 10.1115/1.4062980
• 发表时间: 2023
• 期刊: Journal of Mechanical Design
• 影响因子: 3.3
• 作者: Wang, Zhichao;Melkote, Shreyes;Rosen, David W.
• 通讯作者: Rosen, David W.

Process-aware part retrieval for cyber manufacturing using unsupervised deep learning

使用无监督深度学习进行网络制造的过程感知零件检索

• DOI: 10.1016/j.cirp.2023.03.020
• 发表时间: 2023
• 期刊: CIRP Annals
• 作者: Yan, Xiaoliang;Wang, Zhichao;Bjorni, Jacob;Zhao, Changxuan;Dinar, Mahmoud;Rosen, David;Melkote, Shreyes
• 通讯作者: Melkote, Shreyes