Session: 02-03-01: Session #1: Nanomanufacturing: Novel Processes, Applications, and Process-Property Relationships

Paper Number: 99969

99969 - Towards Secured Process Data Sharing of Metal-Based Additive Manufacturing for Cross-System Part Certification 

Process uncertainty remains a major challenge in the broader adoption of metal-based additive manufacturing (AM) processes. Therefore, it is crucial for users to detect process anomalies in an effective and timely manner and in-situ monitoring and part certification is of critical importance. However, building a robust in-situ process monitoring model is very challenging. Due to the high part complexity and the highly variable part designs and printing parameters in diversified AM applications, large amounts of training data are required for establishing a reliable model for in-situ monitoring and part certification. These obstacles pose serious challenges for AM users, especially those working with small-to-medium sized manufacturers (SMMs).

One potential solution is to aggregate process data from multiple AM users. The aggregated training data can then be leveraged to develop a more accurate, robust, and generalizable machine learning model for anomaly detection. Unfortunately, the major practical obstacle in aggregating process data from multiple AM users is the data privacy concerns that arise from sharing process data outside of the user’s organization. AM process data contains critical product design information, which heavily involves the intellectual property (IP) of the individual user. Sharing these data outside the user’s organization can potentially expose the AM users to the risk of IP theft.

To address these challenges, we first evaluate the data privacy gaps in metal-based AM anomaly detection. Several state-of-the-art feature extraction methodologies for metal-based AM anomaly detection are evaluated for both anomaly detection and privacy preservation. Subsequently, we propose an Adaptive Design De-identification for Additive Manufacturing (ADDAM) methodology for masking the design information in the thermal process data, while simultaneously retaining the quality related information for process monitoring. This methodology will allow for the secure sharing of AM process data among multiple users, which establishes the foundation for data aggregation and transfer learning modeling, leading to the privacy-preserving cross-system anomaly detection models with improved IP security and model robustness. A support vector machine (SVM) classifier is used to evaluate the performance of the proposed ADDAM algorithm in terms of both privacy gain and utility loss. A real-world case study is used to validate the proposed method based on the fabrication of two cylindrical shaped disks using the directed energy deposition (DED) process. The results demonstrate a significant improvement in the privacy preservation related to printing paths, while simultaneously limiting the loss in anomaly detection performance.

Finally, several challenges and research opportunities are discussed for secured data sharing and collaborative part certification in metal-based AM processes.

Presenting Author: Wenmeng Tian Mississippi State University

Presenting Author Biography: Dr. Wenmeng Tian received her Ph.D. degree in Industrial and Systems Engineering from Virginia Tech in 2017. Her research focuses on advanced sensing and analytics for advanced manufacturing process modeling, monitoring, and prognosis. Her research has been applied to both subtractive and additive manufacturing processes. Her publications have appeared in journals such as IISE Transactions, Journal of Manufacturing Science and Engineering, and Additive Manufacturing. Her work has been funded by NSF, DoD, DoL, and industrial institutes. She recently received the NSF CAREER Award in 2021.

Authors:

Wenmeng Tian Mississippi State University