Novel strategy for optimizing of corn starch-based ink food 3D printing process: Printability prediction based on BP-ANN model

Xueyuan Jiao,Guangyue Ren,Chung Lim Law,Linlin Li,Weiwei Cao,Zhenjiang Luo,Lifeng Pan,Xu Duan,Junliang Chen,Wenchao Liu
DOI: https://doi.org/10.1016/j.ijbiomac.2024.133921
Abstract:Although starch has been intensively studied as a raw material for 3D printing, the relationship between several important process parameters in the preparation of starch gels and the printing results is unclear. In this study, the relationship between different processing conditions and the gel printing performance of corn starch was evaluated by printing tests, rheological tests and low-field nuclear magnetic resonance (LF-NMR) tests, and a back-propagation artificial neural network (BP-ANN) model for predicting gel printing performance was developed. The results revealed that starch gels exhibited favorable printing performance when the gelatinization temperature ranged from 75 °C to 85 °C, and the starch content was maintained between 15 % and 20 %. The R2adj of the BP-ANN models were all reached 0.894, which indicated good predictive ability. The results of the study not only provide theoretical support for the application of corn starch gels in 3D food printing, but also present a novel approach for predicting the printing performance of related materials. This method contributes to the optimization of printing parameters, thereby enhancing printing efficiency and quality.
What problem does this paper attempt to address?