Hybrid Serving of DOE and RNN-Based Methods to Optimize and Simulate a Copper Flotation Circuit

Prediction of metallurgical responses during the flotation process is extremely vital to increase efficiency using a proper modeling approach. In this study, two new variants recurrent neural network (RNN) method were used predict copper ore indices, i.e., grade and recovery within different operating conditions. The model input parameters including pulp pH solid content as well frother collector dosages first analysed then optimized two-step factorial statistical analysis showed reliable correlation between with coefficients 99.86% 94.50%, respectively. main effect plots indicated that positively affect while increasing negatively influenced quality final concentrate. Despite same from pH, reverse effects other variables observed for recovery. Process optimization revealed maximum 44.39% 11.48% could be achieved under optimal condition 10, 20%, concentrations 25 g/t 9.9 g/t, Then, predictive long short-term memory (LSTM) gated unit (GRU) networks structure evaluated mean square error (MSE), root (RMSE), absolute percentage (MAPE), coefficient (R2). simulation results LSTM higher R2 0.963 0.934 recovery, respectively, was more effective than GRU algorithm corresponding values 0.956 0.919, show useful in predicting behaviour response changes parameters.