The laser powder bed fusion (LPBF) process is strongly influenced by the characteristics of the powder layer, including its thickness and thermal transport properties. This paper investigates in-situ characterization of the powder layer using thermographic inspection. A thermal camera monitors the temperature history of the powder surface immediately after a layer of new powder is deposited by the recoating system. During this process, thermal energy diffuses from the underlying solid part, eventually raising the temperature of the above powder layer. Guided by 1D modeling of this heat-up process, experiments show how the parameterized thermal history can be correlated with powder layer thickness and its thermal conductivity. A neural network, based on the parameterized thermal history, further improves the correlation after training. It is used to predict the part distortion for an unsupported structure. This method detects serious part distortion several layers before the part breaks through the powder layer and interacts with the recoater. This approach can be automated to prevent catastrophic recoater crashes or abrasion of soft wipers and has the potential to monitor local properties of the powder layer in-situ.


Mechanical and Aerospace Engineering


U.S. Department of Energy, Grant DE-EE0007613

Keywords and Phrases

Additive manufacturing; Infrared thermography; Laser powder bed fusion (LPBF); Neural network; Powder thickness; Recoater crash

International Standard Serial Number (ISSN)


Document Type

Article - Journal

Document Version

Final Version

File Type





© 2023 Elsevier, All rights reserved.

Publication Date

01 Jul 2022