Advanced Simulation of Quartz Flotation Using Micro-Nanobubbles by Hybrid Serving of Historical Data (HD) and Deep Learning (DL) Methods

The present study investigates the optimization and advanced simulation of flotation process coarse particles (–425 + 106) using micro-nanobubbles (MNBs). For this purpose, experiments in presence absence MNBs were performed on quartz particles, results statistically analyzed. Methyl isobutyl carbinol (MIBC) was employed as a frother for generating through hydrodynamic cavitation. significance operating variables, including impeller speed, air flow rate, together with bubble size, particle size recovery assessed historical data (HD) design analysis variance (ANOVA). correlation between parameters response modeled hybrid convolutional neural networks (CNNs) recurrent (RNNs) deep learning (DL) frameworks to automatically extract features from input CNN base layer. ANOVA indicated that all variables affect responses meaningfully. Significant interactions found rate well speed MNB size. It CNN-RNN model could finally be used based intelligent results. Based Pearson coefficients (PCCs), it evident had strong linear relationship recovery. However, Shapley additive explanations (SHAP) considerably more accurate predicting relationships than correlations, even though outputs agreed well.