Machine learning-assisted E-jet printing for manufacturing of organic flexible electronics

Abstract

Electrohydrodynamic-jet (E-jet) printing technique enables the high-resolution printing of complex soft electronic devices. As such, it has an unmatched potential for becoming the conventional technique for printing soft electronic devices. In this study, the electrical conductivity of the E-jet printed circuits was studied as a function of key printing parameters (nozzle speed, ink flow rate, and voltage). The collected experimental dataset was then used to train a machine learning algorithm to establish models capable of predicting the characteristics of the printed circuits in real-time. A decision tree was applied to the data set and resulted in an accuracy of 0.72, and further evaluations showed that pruning the tree increased the accuracy while sensitivity decreased in the highly pruned trees. The k-fold cross-validation (CV) method was used in model selection to test the ability of the model to get trained on data. The accuracy of CV method was the highest for random forest at 0.83 and K-NN model (k = 10) at 0.82. Precision parameters were compared to evaluate the supervised classification models. According to F-measure values, the K-NN model (k = 10) and random forest are the best methods to classify the conductivity of electrodes.