An important challenge for additive manufacturing and 3D printing processes is accurate and repeatable deposition quality. Current approaches are unable to handle variable process parameters and input material quality. Accurately controlling material properties requires predicting material state changes. This work proposes a model using statistical learning techniques in conjunction with iterative material study to identify and compute the sources of defects and local material properties. The model makes use of the element-by-element fabrication and time-series material changes of additive manufacturing. The deposition of a part is segmented into volume elements, called voxels. Each deposited voxel is treated as an independent sample of the process parameter effects. The time series of deposition is treated as a Markov Chain, with the control parameters and measurable emissions as known quantities. The state of the material is a hidden variable. The hidden variable is approximated using material models and post-fabrication testing results to train the distribution embedded in the Markov Chain. The results indicated that a physics-based material state transition matrix in conjunction with final material properties and time-series of physical emissions can give insight into process variability and control errors. These results have wide ranging implications as a computationally efficient means of iterative process improvement for additive manufacturing, designing new control strategies, and revealing the real-time state of voxels as they are deposited. This approach moves closer to a predictive model that includes current information on the state of the process to update the prediction.

This content is only available via PDF.
You do not currently have access to this content.