Optimizing the impact strength and hardness of the liquid crystal display printed parts using artificial neural network

Vat photopolymerization (VPP) is an effective additive manufacturing (AM) process known for its high dimensional accuracy and excellent surface finish. The combination of visible light with the use of LCD screens for 3D printing, allows for a faster, more efficient and economical manufacturing process. Despite these benefits, fabricating the end-use products still has some limitations related to the strength of the fabricated parts. For this purpose, the present paper provides a methodology to predict and optimize three critical process variables in AM, namely: layer height, build orientation, post-curing time. A neural-network model was developed for predicting the impact strength and hardness and optimizing the printing variables for highest responses. From the experiments using full-factorial design, it was revealed that improved parts strength and hardness are obtained at lower layer height, flat orientation, and moderate post-curing time. Based on the ANOVA analysis of, the most effective variable on the impact strength was post-curing time with (41.8%), while the orientation was higher contribution than the rest on the parts hardness with (47.5%). Comparisons between the experimental and the predicted values were illustrated. The mean error between experimental and neural network model was (1.13%) for impact strength and (0.82%) for hardness strength with correlation coefficient equal to 0.988 and 0.982 for the two responses respectively. The current proposed study demonstrates good agreement between the predicted model values and the experiments outcomes of impact strength and parts hardness.