Graphene as an Active Material for Supercapacitors: A Machine Learning Approach

Anif Jamaluddin (ESMART, Department of Physics Education, Universitas Sebelas Maret, Surakarta Center of Excellence for Electrical Energy Storage Technology, Universitas Sebelas Maret, Surakarta)
Annisa Dwi Nursanti (ESMART, Department of Physics Education, Universitas Sebelas Maret, Surakarta)
Anafi Nur'aini (School of Energy Systems, LUT University)
Rekyan Regasari M Putri (Department of Informatic, Brawijaya University, Malang)
Muhammad Usama Arshad (Materials Science and Engineering Department at Texas A&M University, College Station, Texas)

INDONESIAN JOURNAL OF APPLIED PHYSICS
Vol 13, No 2 (2023): October

Graphene is a promising material for supercapacitors due to its unique properties, which influence the device's supercapacitor. This study aims to investigate the key factor of graphene properties in supercapacitors (, with the goal of improving their performance. Also, we observe the machine learning models for predicting capacitance of supercapacitor including four algorithms of machine learning: Linear Regression (LR), lazy IBK, Decision Table (DT), and Random Forest (RF). Machine learning model showed that the RF model demonstrated the highest correlation value of 0.745, surpassing other models. Also, the study revealed that graphene has a high specific surface area and highly porous structure, which enhanced the high capacitance values. Finally, these machine learning models are suitable to apply in materials sciences field for understanding the materials properties in supercapacitor.