PREDICTION OF MECHANICAL PROPERTIES APPLYING A NEURO-FUZZY INFERENCE SYSTEM IN ADDITIVE MANUFACTURING

Abstract

In this work the applicability of an artificial intelligence tool is evaluated for the prediction of mechanical properties in parts built by additive manufacturing (AM). The AM process brings the possibility to process many materials from polymers and ceramics to metals, however the applicability of this technology is limited due to the anisotropy inherent to the layered manufacturing process, which generate porosity between adjacent layers accelerating the degradation of the parts built. For the porosity prediction in samples of stainless steel 316L built by selective laser melting (SLM) a hybrid machine learning tool is applied. A total of 64 data sets were used, of which 80% was used for training, 10% for validation and 10% for prediction. Different hyperparameters configurations were evaluated until predictions were obtained with minimum error, the accuracy of the system was evaluated by applying three statistical metrics: mean square error (RMSE), mean absolute percentage error (MAPE) and the coefficient of determination (R2). In conclusion, it is established that the use of a neuro-fuzzy inference system is easy to implement, and the precision reached is 1,364, 0.129 and 0.9998 for RMSE, MAPE and R2 respectively.

Keywords: Additive manufacturing, Selective laser melting, Porosity, Neuro-fuzzy inference system

References

[1]A. Alafaghani, A. Qattawi, B. Alrawi, and A. Guzman, “Experimental Optimization of Fused Deposition Modelling Processing Parameters: A Design-for-Manufacturing Approach,” Procedia Manuf., vol. 10, pp. 791–803, 2017.

[2]M. Rinaldi, T. Ghidini, F. Cecchini, A. Brandao, and F. Nanni, “Additive layer manufacturing of poly (ether ether ketone) via FDM,” Compos. Part B Eng., vol.
145, no. December 2017, pp. 162–172, 2018.

[3]N. Li et al., “Progress in additive manufacturing on new materials: A review,” J. Mater. Sci. Technol., vol. 35, no. 2, pp. 242–269, 2019.

[4]Y. Zhu, J. Zou, and H. Yang, “Wear performance of metal parts fabricated by selective laser melting: a literature review,” J. Zhejiang Univ. A, vol. 19, no. 2, pp.
95–110, 2018.

[5]C. Y. Yap et al., “Review of selective laser melting:Materials and applications,” Appl. Phys. Rev., vol. 2, no. 4, 2015.

[6]X. Zhang, C. J. Yocom, B. Mao, and Y. Liao, “Microstructure evolution during selective laser melting of metallic materials: A review,” J. Laser Appl., vol. 31,
no. 3, p. 031201, 2019.

[7]J. Zhang, B. Song, Q. Wei, D. Bourell, and Y. Shi, “A review of selective laser melting of aluminum alloys: Processing, microstructure, property and developing
trends,” J. Mater. Sci. Technol., vol. 35, no. 2, pp. 270–284, 2019.

[8]Z. Zhang, B. Chu, L. Wang, and Z. Lu, “Comprehensive effects of placement orientation and scanning angle on mechanical properties and behavior of 316L stainless steel based on the selective laser melting process,” J. Alloys Compd., vol. 791, pp. 166–175, 2019.

[9]O. O. Salman et al., “Impact of the scanning strategy on the mechanical behavior of 316L steel synthesized by selective laser melting,” J. Manuf. Process., vol. 45, no. July, pp. 255–261, 2019.

[10]K. Lin et al., “Selective laser melting processing of 316L stainless steel: effect of microstructural differences along building direction on corrosion behavior,” Int. J. Adv. Manuf. Technol., vol. 104, no. 5–8, pp. 2669–2679, 2019.