E. M. Scheideler, A. Ahlemeyer-Stubbe

Quality Control of Additive Manufacturing using Statistical Prediciton Methods

in: Proceedings 7th International Conference September 28 and 29, 2017
Pordenone, Italy, Production Engineering and Management, edited by
Elio Padoano, Franz-Josef Villmer


Additive Manufacturing (AM) is increasingly used to design new products.
This is possible due to the further development of the AM-processes and
materials. The lack of quality assurance of AM built parts is a key
technological barrier that prevents manufacturers from adopting.

The quality of an additive manufactured part is influenced by more than 50 parameters,
which make process control difficult. Current research deals with using real
time monitoring of the melt pool as feedback control for laser power.
This paper illustrates challenges and opportunities of applying statistical predictive

modeling and unsupervised learning to control additive manufacturing. In
particular, an approach how to build a feedforward controller will be discussed.


Additive manufacturing, Process control, Predictive modeling, Predictive control