Prediction of Key Parameters for Seamless Steel Tube Piercing with Particle Swarm Optimization-based Random Forest Model

A prediction method using a random forest（RF）model optimized by particle swarm optimization（PSO）is proposed to achieve accurate prediction of key parameters in the seamless steel tube piercing process. Firstly,the production characteristic data are preprocessed, including missing value processing,standardization and one-hot encoding for categorical features. Then,random forest model is used to model the multi-output variables,and the random forest hyperparameters are optimized by PSO so as to improve the prediction accuracy of the model. The experimental results show that the PSO-optimized random forest model is superior to other models in the prediction of roll gap,guide plate gap and plug advance, resulting in a significant reduction of mean squared error and mean absolute error, and improvement in coefficient of determination R². The proposed method can provide a scientific decision-making basis for the seamless steel tube piercing process.