Prediction and Optimization of Microhardness of Electroless Ni–P Alloy Coating Based on Machine Learning

This study explores the design and optimization of parameters in electroless nickel-phosphorus alloy coating using machine learning methods to quickly achieve desired hardness values. Three different types of models, including K-Nearest Neighbors, Extreme Gradient Boosting (XGBoost), and Support Vector Regression, were employed to predict coating hardness based on electrolyte composition and process parameters. The relationship between electroless bath parameters and coating hardness was revealed using SHapley Additive exPlanations analysis. The results showed that the XGBoost model outperformed the other two models, achieving an R 2 of 0.924 and an RMSE of 0.0651. The pH value, bath temperature, and sodium citrate concentration have the most significant impact on coating hardness. Finally, three samples were selected for experimental verification, and it was found that the XGBoost model could accurately predict the hardness values. This demonstrates the feasibility and practicality of using machine learning methods to predict the hardness of electroless nickel-phosphorus alloy coatings.