The Prediction Model of SAG Mill Load Based on Industrial Big-data Mining

Regarding the modeling problem of grinding process in mining, taking the operating load of semi-automatic grinding mill as the research object, starting from industrial big data analysis, using data mining methods to establish decision tree models and random forest models, and evaluating and predicting the constructed models. Through testing, decision tree models with different effects can be obtained, indicating that the decision tree model has a certain degree of flexibility, and selecting parameters can lead to a model with good practicality. In addition, under the same conditions, a random forest model was constructed, and simulation results showed that the combination model constructed using the random forest algorithm was superior to a single decision tree model, with better prediction performance.