Deep Learning for Prediction of Print Parameters and Realized Electrical Performance and Geometry on Inkjet Platform

Abstract

A closed-loop deep learning approach for correlating the print parameters with realized electrical performance and geometry estimations on an inkjet platform has been presented in this paper. An estimate of the changes in the print parameters and the recognized print dimension is necessary to print reliable and fine conductive traces. The inks used for this analysis are both particle and particle-free silver inks, and the comparison of the same is also studied. A closed-loop control algorithm is used to attain the desired electrical and geometrical values by changing the print parameters without any user intervention. Sensing is achieved by an automatic print parameter sensing system using a camera that captures the print to identify the geometry and dimension of the same. Once the realized print parameters are determined, a deep learning neural network regression model based on these parameters is used to predict the desired input print parameters, which are used to achieve the desired geometry and dimension of the print. These new parameter values are passed on to the printing software to optimize the print and attain the desired geometry and characteristics.