Presentation Information

[20a-P04-9]Deep Learning of Transparent Conductive Films Fabricated by Carbon Nanotube Dispersions

〇Yusuke Yasuda1, Hirokuni Jintoku1, Shun Muroga1, Hiroshi Morita1, Kenji Hata1 (1.National Institute of Advanced Industrial Science and Technology)

Keywords:

Deep Learning,Transparent Conductive Films,Carbon Nanotube

Deep-learning-based material design has been developed so far however, the application of deep learning methods to materials and devices that undergo multi-step fabrication process has not yet been achieved. Then we focused on creating such a deep learning model, using transparent conductive films made from carbon nanotube (CNT) dispersions as an example. Transparent conductive films are fabricated through dispersion and film-making process. In this study, we aimed to fabricate deep learning model specially for the dispersion step. Optical microscopic images are used as input, and the experimenter-defined dispersity was used as output. The deep learning model was constructed based on convolution neural network (CNN), and the multi-categorical loss function decreases with learning epoch, indicating that learning was processing.