This study explores the application of machine learning techniques for predicting generic mill liner wear in semi-autogenous grinding (SAG) mills used in mineral processing. Various models were developed and compared using data from 143 liner measurements across 36 liner cycles from ten different SAG mills. The research initially focused on individual mill modeling, employing simple linear regression, first-order kinetic approach, Multiple Linear Regression (MLR), tree-based methods (Decision Trees, Random Forests, XGBoost), and Multilayer Perceptron (MLP). Results showed that simple linear regression provided sufficient accuracy, with other methods only slightly improving performance. This study then developed a combined model using data from multiple mills. MLR and advanced machine learning techniques were applied for this generic model, with XGBoost emerging as the most successful. In the interpolation scenario involving a mill similar to those in the training data, the XGBoost model achieved a mean absolute percentage error (MAPE) of 5.27%. For the extrapolation scenario, with a mill larger than those in the training set, the MAPE increased slightly to 6.12%. These results demonstrate the potential of machine learning approaches in creating effective generic models for mill liner wear prediction. However, this study also highlights the potential for improving predictive models by incorporating additional key parameters such as liner and ball material properties.

Recent studies in the flotation of fine particles have necessitated new techniques and analyses for developing various strategies. Particularly, the improvements in flotation chemistry including the selection of the type of frother, collector, and other reagents have become very significant. In this study, the effect of different commercial polypropylene glycol frothers (PPG200, 400, and 600) in the presence of dodecylammonium hydrochloride (DAH) was investigated for their contribution to flotation recoveries and bubble–particle attachment time values of fine quartz minerals. Zeta potential measurements with DAH were also carried out as a function of pH and reagent concentration to justify the effect of collector usage alone on the charge of particles. A linear increase in flotation recoveries against collector concentration, e.g., 7.4% recovery at 1 × 10⁻⁵ mol/L DAH and 65.4% recovery at 1 × 10⁻³ mol/L DAH, was obtained. In this context, the contribution of frothers was particularly important in that a recovery of 15.91% in the absence of the frother and a modest increase to 19.70% was obtained upon the addition of PPG600 at its critical coalescence concentration (CCC) of 3 ppm. Finally, a strong correlation was found between the bubble–particle attachment time and flotation recovery as a function of collector concentration (lowest attachment time vs. highest flotation recovery). The latter correlation is very promising because bubble attachment time leads to various micro-mechanisms in flotation including bubble film thinning, bubble rupture, and induction time, and consequently, frother efficiency in the presence and absence of a collector. As a result, the experimental findings were gathered to achieve a consistent base for further fundamental studies on the application of the synergistic effect of frothers and collectors in the flotation of fine particles.

Soft sensors are mathematical models that estimate the value of a process variable that is difficult or expensive to measure directly. They can be based on first principle models, data-based models, or a combination of both. These models are increasingly used in mineral processing to estimate and...

The conventional flowsheet of coal washing plants includes sequential units where different coal size fractions are washed. Generally, coal washeries include dense medium vessels/drums and cylones for coarse and intermediate sizes as well as spirals and flotation units for fine/ultrafine fractions. Three product-dense medium cyclones (TPDMC) may replace the traditional flowsheet, enabling coals of a wide size range to be treated at once. TPDMC is widely used in the world, particularly in China, and has many potential advantages, yet there is limited data about its performance and simulation. In this study, the performance of a 70 t/h feed capacity TPDMC used for hard coal washing was investigated and its separation behavior was simulated using two different dense medium separation models. Test results revealed that the separation performance of TPDMC was efficient, yet different geometric designs might be required for the treatment of various coal types.

The design of the grinding circuits and the control of the transferring load in the ore preparation plants are of great importance from a technical and economic point of view. The importance of the circulating load for grinding process is well known and stated in the literature. However, there...