Content area
The widespread adoption of digital data management methods for transformative technologies, such as additive manufacturing (AM), within the aerospace industry is impeded by poor interoperability between AM component manufacturing processes. Moreover, data quality may be compromised due to sensor failures or other corruptions. Additionally, massive amounts of data are collected during these processes, often needing to remain accessible for decades. These storage costs can place a significant financial burden on smaller suppliers. This work aims to make digital data management methods more affordable and, therefore, approachable for smaller suppliers.
First, the design and initial implementation of an affordable and adaptable data acquisition and management system for AM processes is presented. This system, initially designed for the WarpSPEE3D cold spray AM machine, leverages flexible composite IDs to create unique identifiers for the machine and its builds. Furthermore, a test-driven framework to ensure correctness and scalability of processed data is detailed. This framework is used to identify current limitations and areas for future optimizations.
Then, an application-focused data imputation framework is introduced to address the problem of data quality in AM sensor readings. This framework was applied to model parameter estimation in two ways: first, the theoretical effects of missing or imputed data on the Cramer-Rao Lower Bound (CRLB) of an unknown scalar parameter were derived and evaluated. Second, an experimental case study estimating the parameters of a mass-spring-damper model under various types of missing data was conducted.
Finally, this thesis addresses the challenge of data storage by introducing a model-based data compression framework designed to reduce AM storage costs more effectively than non-model-based data compression methods. This framework was evaluated using both a physics-based model and a Chebyshev polynomial model using a synthetic mass-spring-damper dataset.