A data-driven approach for evaluation of electrolyte informatics on electrochemical performance of carbon-based electrode materials

Electrolytes are one of the most vital constituents of any electrochemical energy storage device. In this study, we investigated the effect of electrolyte parameters like bare ion size (BIS), hydrated ion size (HIS), electrolyte concentration, and operating temperature on the electrochemical performance of the supercapacitor using a large dataset of 900 data. The impact of different types of electrolytes on the performance of energy storage devices was also investigated. Further, the significance of pore size on electrochemical performance was analyzed. Artificial neural network and Random Forest model provided better accuracy with higher correlation coefficient values (0.851 and 0.852, respectively) and lower root mean square error (RMSE) (0.98 and 1.24, respectively) in comparison to other models. The novelty of this work is the inclusion of electrolyte parameters along with structural and chemical composition and large input dataset for prediction, which enhanced the performance accuracy of designed models compared with other reported works in the literature.