Prediction of electrochemical impedance spectroscopy of high-entropy alloys corrosion by using gradient boosting decision tree

Effects of environmental factors ([Cl-], [SO42-], and pH) on corrosion of CoCrFeAlNi2.1 high-entropy alloy were studied by electrochemical impedance spectroscopy (EIS) and machine learning (ML). Various ML models were employed to predict the impedance data of EIS. It was found that the [SO42-] and pH were positively corrected with -ZIm, whereas the [Cl-] was negatively corrected with -ZIm. The pH was the key feature impacting the corrosion of the HEA. Among various ML models, the performance of Gradient boosting decision tree (GBDT) algorithm for this task was the most excellent. The corresponding R2, MAE and RMSE of GBDT were 0.9491, 2930.24, and 13138.96, respectively.