Search Results (18)

In the mining industry, where equipment breakdowns cause expensive unplanned downtime, operational continuity is paramount. Internet of Things (IoT) technologies have the potential to make predictions; however, most solutions lack a holistic view and mapping of complex system interdependencies. This study presents a comprehensive predictive maintenance (PdM) framework specifically designed for continuous-operation mining environments, with a primary focus on Semi-Autogenous Grinding (SAG) mills. By combining exploratory data analysis, advanced feature engineering, classical machine learning (Gradient Boosting Classifier), and deep learning (LSTM with multiple time-window configurations), the system achieves real-time anomaly detection, root-cause explanation, and failure forecasting up to 48 h in advance (average lead time: 17 h). A four-layer digital twin architecture integrated with Streamlit enables actionable alerts classified as emergency, planned, or preventive interventions. Applied to a one-year dataset comprising 99,854 hourly records from an industrial SAG mill, the hybrid model prevented an estimated 219.5 h of unplanned downtime, yielding substantial economic benefits. The proposed solution is deliberately designed for high adaptability across multiple equipment types and industrial sectors beyond mining. Full article

The shell liner is a core component of Semi-Autogenous Grinding (SAG) mills, suffering severe wear from ore impact and friction, and its shape directly affects grinding efficiency and maintenance costs. In this study, the Finnie wear model in EDEM2022 software was improved to predict the wear morphology evolution of shell liners. A Python-based coupled simulation of the Discrete Element Method (DEM, EDEM) and Finite Element Method (FEM, ABAQUS) was established to analyze liner wear mechanisms, stress states, and mill service performance (wear resistance, grinding efficiency, and stress distribution). The simulated wear profile showed high consistency with laser three-dimensional scanning (LTDS) results, confirming the improved Finnie-DEM model’s effectiveness in reproducing liner wear evolution. Shearing in crushing/grinding zones was the main wear cause, with additional contributions from relative sliding among ore, grinding balls, and liners in grinding/discharge zones. DEM-FEM coupling revealed two circumferential instantaneous wear extremes (Max^a > Max^b) and two lifter wear rate peaks (M^a > M^b). In the grinding zone, liner stress distribution matched wear distribution, with maximum instantaneous stress at characteristic points A and B—stress at A reflects liner impact degree, while stress at B indicates mill ore-crushing capacity. Optimizing flat liner shape adjusted wear rate peaks (M^a, M^b), improving overall liner wear. This optimization significantly affected stresses at A/B and ore normal collision but had little impact on mill energy efficiency. Full article

Predictive maintenance of semi-autogenous grinding (SAG) mills reduces unplanned downtime and improves throughput. This study develops a data-driven prognostic model for production SAG mill using four years of operational data (temperature, voltage, current, motor speed, etc.). We follow a MATLAB (R2025a)-based prognostics and health management (PHM) workflow: data cleaning and synchronization; feature engineering in time and frequency domains (statistical moments, spectral power, bandwidth); normalization and clustering to separate operating regimes; and labeling of run-to-failure sequences for a recurring electrical failure mode. A health indicator is derived by scoring candidate features for monotonicity, trendability, and prognosability and fusing them into a condition index. Using MATLAB Predictive Maintenance Toolbox, we train and validate multiple Remaining Useful Life (RUL) learners including similarity-based, regression, and survival models on run-to-failure histories, selecting the best via cross-validated error and prediction stability. On held-out sets, the selected model forecasts RUL consistent with observed failure dates, providing actionable lead time for maintenance planning. The results highlight the practicality of deploying a PHM pipeline for SAG mills using existing plant data and commercial toolchains. Full article

The semi-autogenous grinding (SAG) mill, renowned for its high efficiency, high production capacity, and low cost, is widely used for crushing and grinding equipment. However, the current understanding of the overall particle behavior influencing its efficiency remains relatively limited, particularly the impact of the shape of SAG mill liners on material behavior. This study employs discrete element method (DEM) simulation technology to investigate the effects of different liner structures on particle trajectories and collision energy, systematically investigating the impact of lifter bars angle, height, and the number of lifter bars on grinding efficiency. The results of single-factor simulations indicate that when the lifter bars height (230 mm) and the number of lifter bars (36) are fixed, the total collision energy dissipation between steel balls and ore, as well as among ore particles, reaches a maximum of 526,069.53 J when the lifter bars angle is 25°. When the lifter bar angle is fixed at 25° and the number of lifter bars is set to 36, the maximum collision energy dissipation of 627,606.06 J occurs at a lifter bars height of 210 mm. When the angle (25°) and height (210 mm) are fixed, the highest energy dissipation of 443,915.37 J is observed with 12 lifter bars. Results from the three-factor, three-level orthogonal experiment reveal that the number of lifter bars exerts the most significant influence on grinding efficiency, followed by the angle and height. The optimal combination is determined to be a 20° angle, 12 lifter bars, and a 210 mm height, resulting in the highest total collision energy dissipation of 700,334 J. This represents an increase of 379,466 J compared to the original SAG mill liner configuration (320,868 J). This research aims to accurately simulate the motion of discrete particles within the mill through DEM simulations, providing a basis for optimizing the operational parameters and structural design of SAG mills. Full article

The unreasonable parameters of mill liner lifter bars will not only decrease the operating rate of the mill and increase electricity consumption but, also, seriously restrict the production capacity of the mill. Therefore, optimizing the parameters of liner lifter bars is helpful to save energy, improve its production capacity, and increase benefits for enterprises. Given the unreasonable parameters of the lifter bars of the semi-autogenous grinding (SAG) mill in a beneficiation plant in Yunnan (China), the distinct element method (DEM) with orthogonal tests was used to conduct simulation, the simulation results demonstrating that the three parameters all had significant influence on the collision energy, with the order of group numbers > angles > heights by the analysis of range and variance, and the optimal parameters combination, with angles of 20°, groups of 12, and heights of 210 mm, was obtained. Then, the lifer bars optimized were applied in industrial tests to verify their effect, and the results illustrated that all of the service life of lifter bars, the operating rate, production capacity, and electricity consumption were significantly improved at 159 days, 92.32%, 54.37 t/h, and 21.45 kW·h/t, respectively. This paper proposes a reference for the similar design and optimization of lifter bars for the other beneficiation plants. Full article

The overload problem in semi-autogenous grinding (SAG) mills is critical in the mining industry, impacting the extraction of valuable metals and overall productivity. Overloads can lead to severe operational issues, including increased wear, reduced grinding efficiency, and unscheduled shutdowns, which result in financial losses. Various strategies have been employed to address SAG mill overload, from real-time monitoring to predictive modeling and machine learning techniques. However, existing methods often lack the integration of domain-specific knowledge, particularly in handling class imbalance within operational data, leading to limitations in predictive accuracy. This paper presents a novel approach that integrates convolutional neural networks (CNNs) with physics-informed neural networks (PINNs), embedding physical laws directly into the model’s loss function. This hybrid methodology captures the complex interactions and nonlinearities inherent in SAG mill operations and leverages domain expertise to enforce physical consistency, ensuring more robust predictions. Incorporating physics-based constraints allows the model to remain sensitive to critical overload conditions while addressing the challenge of imbalanced data. Our method demonstrates a significant enhancement in prediction accuracy through extensive experiments on real-world SAG mill operational data, achieving an F1-score of 94.5%. The results confirm the importance of integrating physics-based knowledge into machine learning models, improving predictive performance, and offering a more interpretable and reliable tool for mill operators. This work sets a new benchmark in the predictive modeling of SAG mill overloads, paving the way for more advanced, physically informed predictive maintenance strategies in the mining industry. Full article

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. Full article

The dynamics of milling circuits, particularly those involving Semi-Autogenous Grinding (SAG) mills, are not adequately studied, despite their critical importance in mineral processing. This paper aims to investigate the dynamic behavior of an SAG mill pebble recycling circuit under varying feed ore conditions, focusing on both uncontrollable parameters (such as ore hardness) and controllable parameters (including circuit layout and pebble crusher configurations). The study is carried out with Simulink dynamic simulations. Our findings reveal several key insights. Firstly, plant designs based solely on static simulations may not be adequate for large or complex circuits, as they fail to account for the dynamic nature of milling processes. Second, incorporating stockpiles after pebble crushing can effectively mitigate the impact of dynamic fluctuations, leading to more stable circuit performance. Third, different circuit layouts can facilitate easier maintenance and operational flexibility. Notably, finer pebble crushing can enhance circuit throughput by 5% to 10%. Full article

The drop ball test (DBT) is a common quality control procedure used in many grinding media manufacturing units to evaluate the quality of manufactured balls. Whilst DBTs have provided reasonable data over many years, the quantitative comparison of the energy that the balls are subjected to during the DBT and in high-impact loading environments such as semi-autogenous grinding (SAG) mills remains a grey area. To that end, DBT experiments were conducted, and the discrete element method (DEM) was used to assess the grinding media collision behaviour and the extent of ball impact loading to determine the impact energy spectra of the ball collisions. The impact energy spectra data obtained were used to quantify the energy that the grinding balls are exposed to in the DBT environment. The results showed that larger balls were exposed to relatively higher energy levels and had a higher probability of fracture than smaller balls. Furthermore, early ball breakage in a grinding environment is mostly attributed to the existence of imperfections or pre-existing defaults within the ball, whilst continuous wear is a gradual consequence that deplete balls in the mill. Full article

The rapidly developing mining industry poses the urgent problem of increasing the energy efficiency of the operation of basic equipment, such as semi-autogenous grinding (SAG) mills. For this purpose, a large number of studies have been carried out on the establishment of optimal operating parameters of the mill, the development of the design of lifters, the rational selection of their materials, etc. However, the dependence of operating parameters on the properties of the ore, the design of the linings and the wear of lifters has not been sufficiently studied. This work analyzes the process of grinding rock in SAG mill and the wear of lifters. The discrete element method (DEM) was used to simulate the grinding of apatite-nepheline ore in a mill using different types of linings and determining the process parameters. It was found that the liners operating in cascade mode were subjected to impact-abrasive wear, while the liners with the cascade mode of operation were subjected predominantly to abrasive wear. At the same time, the results showed an average 40–50% reduction in linear wear. On the basis of modelling results, the service life of lifters was calculated. It is concluded that the Archard model makes it possible to reproduce with sufficient accuracy the wear processes occurring in the mills, taking into account the physical and mechanical properties of the specified materials. The control system design for the grinding process for SAG mills with the use of modern variable frequency drives (VFD) was developed. With the use of the proposed approach, the model predictive control (MPC) was developed to provide recommendations for controlling the optimum speed of the mill drum rotation. Full article

Energy consumption represents a significant operating expense in the mining and minerals industry. Grinding accounts for more than half of the mining sector’s total energy usage, where the semi-autogenous grinding (SAG) circuits are one of the main components. The implementation of control and automation strategies that can achieve production objectives along with energy efficiency is a common goal in concentrator plants. However, designing such controls requires a proper understanding of process dynamics, which are highly complex, coupled, and have non-deterministic components. This complex and non-deterministic nature makes it difficult maintain a set-point for control purposes, and hence operations focus on an optimal control region, which is defined in terms of desirable behavior. This paper investigates the feasibility of employing machine learning models to delineate distinct operational regions within in an SAG mill that can be used in advanced process control implementations to enhance productivity or energy efficiency. For this purpose, two approaches, namely k-means and self-organizing maps, were evaluated. Our results show that it is possible to identify operational regions delimited as clusters with consistent results. Full article

Mineral particle size is an important parameter in the mineral beneficiation process. In industrial processes, the grinding process produces pulp with qualified particle size for subsequent flotation processes. In this paper, a hierarchical intelligent control method for mineral particle size based on machine learning is proposed. In the machine learning layer, artificial intelligence technologies such as long and short memory neural networks (LSTM) and convolution neural networks (CNN) are used to solve the multi-source ore blending prediction and intelligent classification of dry and rainy season conditions, and then the ore-feeding intelligent expert control system and grinding process intelligent expert system are used to coordinate the production of semi-autogenous mill and Ball mill and Hydrocyclone (SAB) process and intelligently adjust the control parameters of DCS layer. This paper presents the practical application of the method in the SAB production process of an international mine to realize automation and intelligence. The process throughput is increased by 6.05%, the power consumption is reduced by 7.25%, and the annual economic benefit has been significantly improved. Full article

Considering the continuous increase in production costs and resource optimization, more than a strategic objective has become imperative in the copper mining industry. In the search to improve the efficiency in the use of resources, the present work develops models of a semi-autogenous grinding (SAG) mill using statistical analysis and machine learning (ML) techniques (regression, decision trees, and artificial neural networks). The hypotheses studied aim to improve the process’s productive indicators, such as production and energy consumption. The simulation of the digital model captures an increase in production of 4.42% as a function of mineral fragmentation, while there is potential to increase production by decreasing the mill rotational speed, which has a decrease in energy consumption of 7.62% for all linear age configurations. Considering the performance of machine learning in the adjustment of complex models such as SAG grinding, the application of these tools in the mineral processing industry has the potential to increase the efficiency of these processes, either by improving production indicators or by saving energy consumption. Finally, the incorporation of these techniques in the aggregate management of processes such as the Mine to Mill paradigm, or the development of models that consider the uncertainty of the explanatory variables, could further increase the performance of productive indicators at the industrial scale. Full article

Semi-autogenous (SAG) mills are widely used grinding equipment, but some ore with critical particle sizes cannot be effectively processed by SAG mills and turned into pebbles. This research aims to analyze and compare the properties of raw ore and pebbles from a zinc- and tin-bearing ore. The results show that the contents of sphalerite, cassiterite, biotite, antigorite, pyroxferroite, ferroactinolite, and ilvaite in the raw ore are higher than those in the pebbles, and that the pebbles have higher contents of hedenbergite, chlorite, epidote, actinolite, etc. Meanwhile, the abrasion and impact resistance of pebbles is greater than that of the raw ore. In addition, the sphalerite is evenly embedded, and the grinding process is regular. Fine cassiterite associated with harder minerals is difficult to dissociate; it is often found in softer or brittle minerals which may be easily ground into ore mud. The cassiterite in the pebbles is associated with hard and brittle hedenbergite and soft chlorite, making it difficult to recover. This research provides a good foundation for evaluating the recovery value of pebbles and improving the productivity of the SAG process. Full article

Multiple-input and multiple-output (MIMO) systems can be found in many industrial processes, including mining processes. In practice, these systems are difficult to control due to the interactions of their input variables and the inherent uncertainty of industrial processes. Depending on the interactions in the MIMO process, different control strategies can be implemented to achieve the desired performance. Among these strategies is the use of a decentralized structure that considers several subsystems and for which a SISO controller can be designed. In this study, a methodology based on global sensitivity analysis (GSA) to design decentralized control structures for industrial processes under uncertainty is presented. GSA has not yet been applied for this purpose in process control; it allows us to understand the dynamic behavior of systems under uncertainty in a broad value range, unlike approaches proposed in the literature. The proposed GSA is based on the Sobol method, which provides sensitivity indices used as interaction measures to establish the input–output pairing for MIMO systems. Two case studies based on a semi-autogenous grinding (SAG) mill and a solvent extraction (SX) plant are presented to demonstrate the applicability of the proposed methodology. The results indicate that the methodology allows the design of 2 × 2 and 3 × 3 decentralized control structures for the SAG mill and SX plant, respectively, which exhibit good performance compared to MPC. For example, for the SAG mill, the determined pairings were fresh ore flux/fraction of mill filling and power consumption/percentage of critical speed. Full article