[20p-C32-1~11] 16.3 シリコン系太陽電池

15:00 〜 15:15英語発表

[20p-C32-8] Application of a Machine Learning Method, Random Forest, to the Deposition Conditions of Doped Amorphous Silicon Films

〇CHENXI LI¹, Huynh Thi Cam Tu¹, Keisuke Ohdaira¹ (1.JAIST)

キーワード：

amorphous silicon、machine learning、random forest

This study explores the use of the Random Forest (RF) algorithm within the ORANGE open-source data mining tool to predict deposition conditions for achieving high conductivity in amorphous silicon (a-Si) films. The model utilized k-fold cross-validation and Spearman correlations (SC) to enhance predictive accuracy. We demonstrated that the RF algorithm can reasonably predict the experimental conditions required for higher conductivity in both n-type and p-type a-Si films, offering significant potential for future advancements in the field.