Improvement of forming accuracy in prototyping of solid oxide fuel cells using a photocurable 3D printer

In this study, the objective is to fabricate ceramic components of a solid oxide fuel cell (SOFC) using a photocurable 3D printer, and to improve dimensional accuracy through mathematical optimization of forming parameters. Zirconia models of cubes and cylinders with design sizes ranging from 300 to 1000 μm were fabricated. Experimental data were collected by varying exposure time and light-off delay, while other printing conditions were held constant. Based on these data, second-order multivariable polynomial regression models were constructed to predict the fabricated dimensions. A loss function was defined as the squared difference between the predicted and target dimensions (diameter or edge length), and the optimal parameter settings were identified using the L-BFGS-B algorithm. By applying this optimization method, improvements in the formed dimensions of flat-tube SOFCs have been achieved. These results demonstrate the effectiveness of combining regression modeling and numerical optimization to enhance fabrication precision in ceramic additive manufacturing.