fig2

Figure 2. (A) Schematic diagram of the research process of screening anode of SOEC using ML models; (B) Comparison of actual and predicted values by the model, and the residuals of each predicted value by comparing with the actual values in XGBoost model; (C) General schematic diagram of the gradient boosting decision tree algorithm; (D) Part of the predicted hydrated proton concentration results. SOEC: Solid oxide electrolysis cell; ML: machine learning.