Neuro-fuzzy modelling methods for relative density prediction of stainless steel 316L metal parts produced by additive manufacturing technique

Toprak C. B., DOĞRUER C. U.

Journal of Mechanical Science and Technology, vol.37, no.1, pp.107-118, 2023 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 37 Issue: 1
  • Publication Date: 2023
  • Doi Number: 10.1007/s12206-022-1211-6
  • Journal Name: Journal of Mechanical Science and Technology
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Compendex, INSPEC, Metadex, Civil Engineering Abstracts
  • Page Numbers: pp.107-118
  • Keywords: Additive manufacturing, Artificial neural networks, Machine learning, Neuro-fuzzy modelling, Selective laser melting
  • Hacettepe University Affiliated: Yes


© 2023, The Korean Society of Mechanical Engineers and Springer-Verlag GmbH Germany, part of Springer Nature.Two machine learning (ML) methods, adaptive neuro fuzzy inference system (ANFIS) and artificial neural networks (ANN) have been implemented to predict the relative density (RD) of stainless steel 316L parts which are produced in additive manufacturing (AM) machines. The objective of this paper was to create ML models adapted for AM technique to verify the generalized model that predicts RD with the least error. Some important process parameters in AM such as scanning speed, laser power, hatch distance and layer thickness were picked as input and RD was set as output. Effects of the input parameters on RD were discussed and they were represented in the form of surface plots. It has been found that ANN method’s convergence rate was better than that of ANFIS method, which confirms that usage of neural networks is a better choice than the usage of fuzzy reasoning in modelling AM technique.