Detection of keyhole pore formations in laser powder-bed fusion using acoustic process monitoring measurements

• DOI: 10.1016/j.addma.2022.102735
• Publication Date: 2022-03-31T11:58:38Z
• Authors: Joshua R. Tempelman; Adam J. Wachtor; Eric B. Flynn; Phillip J. Depond; Jean-Baptiste Forien; Gabe M. Guss; Nicholas P. Calta; Manyalibo J. Matt...
• Keywords: Keyhole; Feature (linguistics); Feature vector

In-situ process monitoring of additively manufactured parts has become a topic increasing interest to the manufacturing community. In this work, acoustic measurements recorded during laser powder-bed fusion (L-PBF) were used detect onset keyhole pores induced by lasing process. Post-build radiography was identify locations in build. The pore spatially and temporally registered with time-series position pressure specific partitions signals which correspond formation. Ensemble empirical mode decomposition, traditional Fourier statistical measures corresponding frequency spectra extract feature vectors correlate Sequential selection revealed that associated most useful for identifying formations L-PBF. A subset informative data features train support vector machine model predict formation up 97% accuracy.