CAREER: Smart Additive Manufacturing - Fundamental Research in Sensing, Data Science,and Modeling Toward Zero Part Defects.

职业：智能增材制造 - 传感、数据科学和零件零缺陷建模的基础研究。

• 批准号: 1752069
• 负责人: Prahalada Rao
• 金额: $ 50万
• 依托单位: University of Nebraska-Lincoln
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-04-01 至 2023-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1752069&HistoricalAwards=false
• 关键词: CAREER; Smart; Additive; Manufacturing; Fundamental

Smart Manufacturing strives to monitor every aspect of the manufacturing enterprise - from the individual machine-level to the factory-level - using data gathered from multiple sensors. Resulting efficiencies can reduce product defects and manufacturing costs by over 25 percent. When coupled with Additive Manufacturing, Smart Manufacturing promises to transform U.S. industry. For example, 20 pounds of raw material are currently required to make a one-pound part for the aerospace industry using subtractive machining. Additive Manufacturing can reduce this so-called buy-to-fly ratio of 20:1 to 2:1, while simultaneously reducing lead time from six months to one week. Realization of these potential manufacturing gains will advance the national prosperity and welfare by increasing U.S. advanced manufacturing competitiveness. Despite these advantages, industries are hesitant to adopt Additive Manufacturing due to process inconsistency - parts may have undetected defects, such as porosity, that make them unsafe for use in mission-critical applications. A potential solution to this problem is a concept called Smart Additive Manufacturing, which melds the ideas of Smart Manufacturing with Additive Manufacturing. Through this Faculty Early Career Development Program (CAREER) award, in-process sensor data will be utilized to understand the mechanisms of defect formation occurring during the Laser Powder Bed Fusion Additive Manufacturing process. Advanced data analysis approaches that incorporate the new fundamental understanding of defect evolution will be leveraged to realize a robust correct-as-you-build methodology. This foundational work will find application across many manufacturing sectors including aerospace and defense. The award will also facilitate a discovery-based learning approach to engage learners in hands-on exploration of Additive Manufacturing at multiple levels. A research collaboration with Navajo Technical University will be initiated to further broaden project impact and train the advanced manufacturing workforce of the future. The research goal of this project is to establish a Smart Additive Manufacturing framework for alleviating the poor part quality in the Laser Powder Bed Fusion-based Additive Manufacturing of metals. Success will result in a hybrid Additive Manufacturing strategy that combines material deposition (additive) and material removal (subtractive) actions within the same machine potentially giving rise to zero-defect parts. The research challenges addressed by this award include: 1) understanding how and why certain defects are formed by isolating and quantifying the underlying process phenomena as they happen in real-time using in-process sensors, 2) advancing the mathematics of spectral graph theory to capture defects from heterogeneous sensors in real-time - a big data problem, and 3) forwarding reduced-order models to understand the physical thermomechanical dynamics, such as layer re-melting and reflow, that occur when defects are corrected with hybrid Laser Powder Bed Fusion.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.

智能制造致力于使用从多个传感器收集的数据来监控制造企业的各个方面 - 从单个机器级别到工厂级别。由此产生的效率可将产品缺陷和制造成本减少 25% 以上。智能制造与增材制造相结合，有望改变美国工业。例如，目前使用减材加工为航空航天工业制造一磅重的零件需要 20 磅原材料。增材制造可以将这种所谓的“买飞比”从 20:1 降低到 2:1，同时将交货时间从六个月缩短到一周。实现这些潜在的制造业收益将通过提高美国先进制造业的竞争力来促进国家繁荣和福利。尽管有这些优势，但由于工艺不一致，各行业对采用增材制造仍犹豫不决——零件可能存在未检测到的缺陷，例如孔隙率，这使得它们在关键任务应用中使用不安全。这个问题的一个潜在解决方案是智能增材制造的概念，它将智能制造与增材制造的思想融合在一起。通过该学院早期职业发展计划（CAREER）奖，将利用过程中的传感器数据来了解激光粉床融合增材制造过程中发生缺陷形成的机制。将利用结合了对缺陷演变的新基本理解的高级数据分析方法来实现强大的“即建即正确”方法。这项基础工作将在包括航空航天和国防在内的许多制造领域得到应用。该奖项还将促进基于发现的学习方法，让学习者在多个层面上对增材制造进行实践探索。将启动与纳瓦霍技术大学的研究合作，以进一步扩大项目影响并培训未来的先进制造劳动力。该项目的研究目标是建立一个智能增材制造框架，以缓解基于激光粉床熔融的金属增材制造中零件质量差的问题。成功将带来混合增材制造策略，该策略将材料沉积（增材）和材料去除（减材）操作结合在同一台机器内，有可能产生零缺陷零件。该奖项解决的研究挑战包括：1) 通过使用过程传感器实时隔离和量化潜在的过程现象，了解某些缺陷是如何以及为何形成的，2) 推进谱图理论的数学发展实时捕获来自异构传感器的缺陷 - 这是一个大数据问题，3）转发降阶模型以了解物理热机械动力学，例如使用混合激光粉末床校正缺陷时发生的层重熔和回流融合.This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。

项目成果

Defect Detection and Monitoring in Metal Additive Manufactured Parts through Deep Learning of Spatially Resolved Acoustic Spectroscopy Signals

通过空间分辨声谱信号的深度学习来检测和监测金属增材制造零件中的缺陷

• DOI: 10.1520/ssms20180035
• 发表时间: 2018-11
• 期刊: Smart and Sustainable Manufacturing Systems
• 影响因子: 1
• 作者: Williams, Jacob;Dryburgh, Paul;Clare, Adam;Rao, Prahalada;Samal, Ashok
• 通讯作者: Samal, Ashok

Monitoring and prediction of porosity in laser powder bed fusion using physics-informed meltpool signatures and machine learning

使用物理信息熔池特征和机器学习监测和预测激光粉末床熔合中的孔隙率

• DOI: 10.1016/j.jmatprotec.2022.117550
• 发表时间: 2022-06
• 期刊: Journal of Materials Processing Technology
• 影响因子: 6.3
• 作者: Smoqi, Ziyad;Gaikwad, Aniruddha;Bevans, Benjamin;Kobir, Md Humaun;Craig, James;Abul;Peralta, Alonso;Rao, Prahalada
• 通讯作者: Rao, Prahalada

Thermal Modeling in Metal Additive Manufacturing Using Graph Theory: Experimental Validation With In-Situ Infrared Thermography Data From Laser Powder Bed Fusion

使用图论进行金属增材制造中的热建模：利用激光粉末床熔合的原位红外热成像数据进行实验验证

• DOI: 10.1115/msec2020-8433
• 发表时间: 2020-09
• 期刊: ASME Manufacturing Science and Engineering Conference
• 作者: Yavari, Reza;Williams, Richard;Cole, Kevin;Hooper, Paul;Rao, Prahalad
• 通讯作者: Rao, Prahalad

Flaw Detection in Wire Arc Additive Manufacturing Using In-Situ Acoustic Sensing and Graph Signal Analysis

使用原位声学传感和图形信号分析进行电弧增材制造中的缺陷检测

• DOI: 10.1115/msec2023-101622
• 发表时间: 2023-06
• 期刊: American Society of Mechanical Engineers
• 作者: Bevans, Benjamin;Ramalho, André;Smoqi, Ziyad;Gaikwad, Aniruddha;Santos, Telmo G.;Rao, Prahalad;Oliveira, J. P.
• 通讯作者: Oliveira, J. P.

In-process monitoring and prediction of droplet quality in droplet-on-demand liquid metal jetting additive manufacturing using machine learning

使用机器学习对按需液滴液态金属喷射增材制造中的液滴质量进行过程监控和预测

• DOI: 10.1007/s10845-022-01977-2
• 发表时间: 2022-06-26
• 期刊: Journal of Intelligent Manufacturing
• 影响因子: 8.3
• 作者: A. Gaikwad;Tammy Chang;B. Giera;N. Watkins;S. Mukherjee;A. Pascall;D. Stobbe;Prahalada K. Rao
• 通讯作者: Prahalada K. Rao