Semi-autogenous Mill Research Articles

Multivariable predictive models for the estimation of power consumption (kW) of a Semi-autogenous mill applying Machine Learning algorithms [Modelos predictivos multivariables para la estimación de consumo de potencia (kW) de un molino Semi - autógeno aplicando algoritmos de Machine Learning

This research aimed to develop machine learning (ML) models to estimate power consumption (Kw) in a Semi-autogenous mill in the mining industry. Using Machine Learning algorithms considering various operating variables for the different models such as Multiple Linear Regression (RLM), Decision Tree Regression (RAD), Random Forest Regression (RBA) and Regression Artificial Neural Networks (ANN). The methodology adopted was applied, with an experimental design with a descriptive and transversal approach. The results of the application of these models revealed significant differences in terms of predictive efficiency. The RLM and RRNA stood out with coefficients of determination (R²) of 0.922 and 0.939, respectively, indicating a substantial capacity to explain the variability in power consumption. In contrast, the tree-based models (RAD and RBA) showed inferior performance, with R² of 0.762 and 0.471. When analyzing key metrics such as Mean Absolute Error (MAE), Mean Square Error (MSE) and Root Root Mean Square Error (RMSE), it was confirmed that both RLM and RRNA outperformed the tree-based models. These results support the choice of RLM and RRNA as preferred models for estimating power consumption in a Semi-autogenous mill.

Scale-Up Investigation of a Pilot and Industrial Scale Semi-Autogenous Mill Using a Particle Scale Model

A particle scale model based on a full two-way coupling of the Discrete Element Method (DEM) and Smoothed Particle Hydrodynamics (SPHs) methods is applied to SAG mills. Motion and collisions of resolved coarser particles within an SAG mill are performed by the DEM component. Fine particles in the feed combine with the water to form a slurry, which is represented by the SPH component of the model. Slurry rheology is controlled by solid loading and fine particle size distribution for each volume of slurry. Transport, dispersion, and grinding of the slurry phase particle size distribution are predicted by solving additional coupled advection–diffusion equations in the SPH component of the model. Grinding of the finer particles in the slurry due to collisions and shear of the coarser particles (rocks and grinding media) is achieved via the inclusion of population balance terms in these equations for each SPH particle. This allows prediction of the transport of both coarser and finer material within the grinding and pulp chambers of an SAG mill, including the discharge performance of the mill. This particle-scale model is used to investigate the relative performance (throughput, product size distribution, resident particle size distribution, net power draw, wear) for an SAG mill at a pilot scale and a 36 ft industrial scale. The 36′ SAG mill considered is a geometrically scaled-up version of the 1.8 m Hardinge pilot scale mill but with a longer belly length, reflecting current SAG mill design preferences. The belly lifters are scaled to a lesser degree with a larger number of lifters used (but still many fewer liners than would typically be used in a large SAG mill based on conventional liner selection rules). The model shows that despite reasonable qualitative similarities, many aspects of the charge structure, slurry transport, coarse particle and slurry discharge through the grates, and the collision energy spectra vary in important ways. This demonstrates that a near purely geometric scale-up of an SAG mill is not sufficient to produce a comparable performance at the two physical mill scales.

Advances in vibration analysis and modeling of large rotating mechanical equipment in mining arena: A review

Large-scale rotating mechanical equipment in the mining arena plays a pivotal role in mining production, where vibration issues directly influence production efficiency and safety. This Review aims to provide a comprehensive review of the latest advancements and methodologies related to the generation mechanisms, identification, and applications of vibrational characteristics in large-scale mining rotating mechanical equipment. Semi-autogenous mills, ball mills, and coal mills are selected as archetype equipment, and the Lagrangian motion equation is employed to unveil the generation mechanisms of vibrations and the embedded physical information in the signals of these machines. Initially, the research delves deeply into the acquisition, extraction, and identification of vibrational signal features, emphasizing that while mechanical vibration signals can reveal the internal operational state and fault information of machinery, there remains a need to enhance their capability to depict complex vibrational signals. Subsequently, this Review discusses in depth the studies focused on predicting the vibrational state of equipment by establishing accurate and reliable soft measurement models, pointing out that current models still have room for improvement in prediction accuracy and generalization capabilities. Conclusively, based on the elucidation of mechanical vibration mechanisms and the collation and outlook of the existing research study, the importance of on-site monitoring, deep learning, Internet of Things technology, and full lifecycle management is accentuated. To better support practical engineering applications, further exploration into the physical properties of vibrational signals and the mechanisms of mechanical vibrations is essential.

Hierarchical Intelligent Control Method for Mineral Particle Size Based on Machine Learning

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.

Fusion-based color and depth image segmentation method for rocks on conveyor belt

Semi-autogenous mill (SAG) feeding process requires accurate and reliable monitoring and controlling techniques for multi-sized rock distribution to stabilize crushing productivity. This study presents a color and depth (RGB-D) image fusion algorithm for the segmentation of multiscale rock images from the conveyor belt. The algorithm combines the multiscale characteristics of the color image and depth image, considers the extraction of large particles from the depth image and the rich texture details for segmentation of the color image, divides the segmentation regions according to different particle sizes, and finally obtains the contour of each particle after region-based superpixels merging. The proposed method outperforms any single sensor image segmentation method proposed in direct comparison, thereby attaining the closest to ground truth performance. The comparison results also show that the accuracy and adaptability of the algorithm are significantly improved, and this algorithm can therefore be used in rock-like material segmentation with multiple size distribution on conveyors.

Acoustic Sensing and Supervised Machine Learning for In Situ Classification of Semi-Autogenous (SAG) Mill Feed Size Fractions Using Different Feature Extraction Techniques

The harsh and hostile internal environment of semi-autogenous (SAG) mills renders real-time monitoring of some critical variables practically unmeasured. Typically, feed size fractions are known to cause mill fluctuations and impede the consistent processing behaviour of ores. There is, therefore, the need for continuous monitoring of mill parameters for optimal operation. In this paper, an acoustic-based sensing method is employed to estimate, in real time, a snapshot of the different feed size fractions presented to a laboratory-scale SAG mill. Employing the MATLAB 2020b programme, the mill acoustic signal is processed using various transform techniques such as power spectral density estimate (PSDE) by Welch’s method, discrete wavelet transform (DWT), wavelet packet transform (WPT), empirical mode decomposition (EMD), and variational mode decomposition (VMD). Different fractional bandpowers are obtained from the PSDE spectrum, while the statistical root mean square values are further extracted from DWT, WPT, EMD, and VMD as feature vectors. The features are used as input features in different machine-learning classification algorithms for different mill feed size fractions predictions. The various transform techniques and feed size fraction predictions are evaluated using the various performance indicators obtained from the confusion matrix such as accuracy, precision, sensitivity and F1 score. The study showed that the acoustic signal feature extraction techniques used in conjunction with the Support Vector Machine (SVM), linear discriminant analysis (LDA), and ensemble with subclass discriminant machine learning algorithms demonstrated improved performance for predicting feed size variations.

Study on impact dynamic characteristics of semi-autogenous mill liner based on improved multivariable method

The impact resistance of semi-autogenous mill liner has an important influence on the reliable operation of the system. The complex dynamic characteristics in the process of collision with abrasive medium are studied by the improved multivariable method proposed in this paper. Firstly, based on manifold element method, the deformation of abrasive medium and lining plate in the contact area is expressed by the coordinates of first-order manifold element basis function, while the deformation of the non-contact region is described by modal coordinates to improve the traditional multivariable method, and the displacement constraint equations as well as the acceleration constraint equations of the collision system are derived, then the Lagrange method is used to establish the impact dynamic equation of abrasive medium and lining plate system. Secondly, a collision test rig between abrasive medium and lining plate was built to verify the correctness and accuracy of the proposed method. Finally, the influence of structural parameters and motion parameters of abrasive medium and lining plate on the dynamic characteristics of collision system is analyzed by parameter analysis method. It is found that when the Poisson’s ratio of abrasive medium to lining plate material exceeds 1.0, the increase amplitude of stress becomes smaller. The increase amplitude of displacement, velocity, acceleration and stress at the collision contact point decreases when the initial velocity of abrasive medium exceeds 3.0 m/s. The stress at the collision contact point changes greatly, and the maximum acceleration occurs in advance when the radius of abrasive medium exceeds 20 mm. When the thickness of liner exceeds 40 mm, the time of maximum acceleration to reach is almost unchanged. The results show that compared with the traditional multivariable method, the improved multivariable method can increase the modeling efficiency and solving accuracy, and lay a theoretical foundation for optimizing the structure of the liner and raising the service life of the system.

Investigation of the effects of rock dynamic hardness on the energy consumption in the crushing unit, case study: Midouk copper mine

ABSTRACT The mounted semi autogenous (SAG) mill in the processing unit of the Midouk copper mine which consumes 35% of operational costs has a critical size within the range of 2.5–5 cm. To optimize the energy consumption in the milling unit of the mine, the data have been collected from 36 blast blocks. These blocks are selected in the main ore body area to monitor their effects on energy consumption in the milling unit. Owing to the dependence of mill energy consumption on the feed strength properties, a criterion for mill performance (MPC) has been defined and compared with the obtained dynamic hardness of seven specific blasting blocks. The results show a linear relationship with the accuracy of R 2 = 0.91 between the dynamic hardness and the MPC. For mines that use SAG or AG mills, the energy consumption for crushing the unit weight of feed depends on their dynamic stiffness.

Differences in Properties between Pebbles and Raw Ore from a SAG Mill at a Zinc, Tin-Bearing Mine

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.

Research on performance of a laboratory-scale SAG mill based on DEM-EMBD

To explore the performance of a semi-autogenous (SAG) mill under different working conditions, in this paper, the working process of the laboratory-scale SAG mill is taken as the research object. Firstly, the electrical system and multi-body dynamics model (EMBD) of the SAG mill are established, the rock discrete element method (DEM) analysis model is obtained based on the multi-sphere filling method, and then the DEM-EMBD coupling model is created, and its effectiveness is verified through actual experiments. Secondly, the influences of fraction of critical speed, lifter face angle, and rock shape on particle motion characteristics, particle collision energy, and mill torque are studied by the method of DEM-EMBD coupling method. The results show that the particle motion characteristics are mainly determined by fraction of critical speed and lifter face angle. Particle collision energy is mainly determined by lifter face angle and rock shape. Further, when the SAG mill runs at a low speed, the lifter face angle and rock shape become the main influencing factors of the mill torque, and the fraction of critical speed becomes the dominant factor with a high running speed of the SAG mill. Finally, the electromechanical characteristics of the SAG mill are analysed, and the results show that the fluctuation of motor load torque, current, and SAG mill speed in the start-up stage is much larger than that in the stable running stage. The fraction of critical speed and the lifter face angle have a greater impact on the electromechanical characteristics, while the rock shape has little influence on the electromechanical characteristics as a whole.

Keys to best practice comminution

Comminution circuits are composed of multiple unit operation, with the objective of reducing mined rock to a size where valuable minerals grains are liberated from gangue. Optimal comminution is critical to achieve efficient mineral separation. There are three reasons for writing this paper. For operators to manage good operations; for designers to produce workable designs; and for educators to provide useful education for mineral process engineers. In all cases, understanding of the transfer size (T80) between the SAG mill and the ball mill is critical to achieve best economics in a semi-autogenous mill (SAG) grinding plant. T80 is important to operators because when the SAG energy and the Bond Ball Mill Work Index on SAG ground ore are measured, accurate prediction of future throughput in any SAG circuit is possible. Without knowledge of the plant T80, it can take many months to figure out how to correct what is really a SAG mill grinding problem, because that problem is hidden if the T80 is not measured. Best Practice Comminution means running a SAG mill under optimal conditions, and avoiding overloading, overspeeding and using excessive steel additions, both during the design and operating stages of the plant. When normal limits for these parameters are exceeded in the design stage, production shortfalls will result, and operating costs will be high. While extra SAG mill capacity is a bonus, lack of capacity is a disaster. This paper shows how to design workable grinding circuits on a particular ore, using either single stage SAG milling, SAB grinding, SABC grinding, or HPGR pre-crushing followed by ball milling. There are many ways to set up a SAG plant, and future expansion should always be considered at the design stage. This opportunity can be overlooked if the designer does not understand the options available.

DETERMINATION OF THE RELATIONSHIP BETWEEN THE WEAR RATE OF GRINDING BALLS IN THE SAG MILL MILL AND THE PARAMETERS THAT CONTROL THE PROCESS

Since the end of the 20th century and the beginning of the 21st century, as a result of the large-scale implementation of the industrialization policy in Azerbaijan, a large number of industrial enterprises with powerful potential have been created. Based on this, the volume of industrial production over the past 10-15 years has increased 2.7-3.0 times. The current state of the country's economy has created ample opportunities for the development of the non-oil sectors along with the oil sector. At the same time, attention is drawn to the variety and speed of work on the extraction of precious metals. Azerbaijan International Mining Company (AIMC) is one such company and is currently engaged in exploration, production and processing at the Gadabay deposit. The high technology used here and the associated equipment and tools emphasize the need to apply and update recovery technologies, as well as various improvements over time. One of the main directions of scientific and practical work carried out here is to increase the resistance of grinding balls to corrosion. The article is devoted to determining the relationship between the wear rate of grinding balls of a semi-autogenous grinding mill and the parameters that control the process. Keywords: industrialization, industrial production, non-oil sector, mining company, drum mill, grinding balls, wear, the relationship between process controllers.

Research on Non-Uniform Wear of Liner in SAG Mill

The numerical calculation method is used to analyze the wear of the liner of the general structure of a semi-autogenous mill in the axial direction, and the non-uniform wear of each area of the liner is studied to explore the reasons for said wear. The liner is divided into areas along the axial direction, and the discrete element method (DEM) is used to analyze the relationship between the wear volume of each area and the total mass of particles. The composition ratio of the rocks and steel balls in each area, and its relationship with time, are also studied. The results show that the total mass of the particles in the area has a significant effect on the wear of the liner. When the particles are affected by the conical end cover on both sides during the operation of the mill, they will be stratified along the axial direction. The particles with large masses will accumulate on both sides of the mill, and the particles with small masses will be concentrated in the middle of the mill. As a result, the difference between the density and impact energy of rocks and steel balls in each area is caused, and eventually, the mill liner appears to have non-uniform wear.

A Dynamic Model for a Class of Semi-Autogenous Mill Systems

In this paper, we develop a state space model of semi-autogenous (SAG) mill systems, which is of great significance to the mineral processing control. According to the practical enterprise production condition, the model is composed of two main parts. One is the ore circuit model. Compared with other models of ore circuit, the present model is simple and efficient. Another part is the SAG subsystem model where the state space model is utilized to depict the grinding mechanism. Furthermore, the grinding process of hold-ups are analyzed and presented with the common outputs such as power draw, pulp density and pulp flow. We consider the mass of rocks, granules and water as states and analyze the observability of the SAG subsystem model. For the ore circuit system and the SAG subsystem established respectively, unknown parameters of these models are identified by appropriate parameter estimation methods. Verified by the actual plant data, the result shows that the SAG mill system works well.

Application of Multi-dimensional Linear Fitting Method in the Establishment of the Semi-autogenous Grinding Mill Model

In order to build the semi-autogenous grinding mill model more accurately and efficiently, a large-scale concentrating mill in China was studied and its daily production data were analyzed. By combining the field experts experience and utilizing the multi-dimensional linear fitting methods, the semi-autogenous grinding mill axial pressure prediction model was built through various methods. Meanwhile, the advantages and disadvantages of different multi-dimensional linear fitting methods were compared to determine the most appropriate method. Then the multi-dimensional linear fitting model for the axial pressure of semi-autogenous mill was successfully achieved, via analyzing the influence of different parameters on the axial pressure model of semi-autogenous mill, and comparing the predicted results of the model with the measured values of the field production. It is demonstrated that this method has high accuracy and calculation speed simultaneously, which is significant to the establishment of semi-autogenous grinding model.

Relationship among operational parameters, ore characteristics, and product shape properties in an industrial SAG mill

A detailed test work was carried out with an industrial semi-autogenous (SAG) mill in a copper plant in close and open circuits to specify the effects of ore characteristics and operational parameters on mill performance. In addition, the shape of mill product particles was studied to monitor breakage events as a function of ore strength. For this purpose, the strength properties of the SAG mill feed and product were specified by different techniques, including the point load index, impact breakage parameters (A × b), abrasion index (ta), and Bond ball mill work index (Wi). Experimental studies showed that an increase in ore hardness is linked with an increase in point load index and Bond work index and a decline in impact breakage parameter (A × b). Hard ore grinding resulted in finer products, compared to the soft ore milling. The softer ore is ground chiefly via the impact breakage whereas the stronger ore is subjected to abrasion and attrition events where the media impact could not sufficiently break the stronger particles. The main reason for the decrease of circuit throughput was changes in breakage mechanisms. The results displayed that the highest aspect ratio and the lowest circularity of mill product are achieved by soft ore milling. As ore strength changes, differences in shape can be attributed to the different breakage mechanisms operated in the mill. In addition, the particle shape characteristics provide a convenient way to quantify changes in SAG mill as a function of the ore properties and the kind of breakage action used.

Dynamic modeling and simulation of a SAG mill-pebble crusher circuit by controlling crusher operational parameters

A semi-autogenous (SAG) or autogenous mills (AG) circuit commonly comprises a pebble (cone) crusher to deal with the critical size particles. The pebbles crusher is often oversized, and thus not fully utilized. The hypothesis put forward in this paper is that active control of the cone crusher’s eccentric speed and closed side setting (CSS) can increase throughput and/or energy efficiency of the SAG circuit. To verify this hypothesis, a dynamic model of a SAG mill-pebble crusher circuit was implemented. The dynamic unit models are based on results presented by several different researchers, whose results here have been combined and implemented in Matlab/Simulink.Previous research shows that by controlling the cone crusher operational parameters, the product particle size distribution (PSD) can be altered during operation. However, few papers have been published about whether the changes of the pebbles crusher’s product PSD, lead to a higher performance of the SAG circuit. This paper examines the relations between the SAG mill and the pebble crusher in closed circuit while putting emphasis on controlling the pebble crusher in order to optimize the circuit throughput.

Prediction model to estimate the specific energy and product particle size distribution of semi-autogenous mill

The optimal use of energy for grinding of ores in mills is one of the key objectives of this study. An attempt has been made to present an experimental model for the specific energy and product particle size distribution of semi-autogenous mills for the simulating of real conditions of the particles collision. A novel semi-autogenous mill test has been suggested in order to measure the required energy for grinding. The effect of the mill’s operation time and various energy levels on the particle size distribution and feed ore hardness has been investigated with samples collected from the Sarcheshmeh Copper Complex, Kerman, Iran. The obtained results demonstrated that there is a high congruence between the experimental product particle size distribution and energy, which are predicted by the proposed novel method and ball-drop test. Using the proposed analytic relationships, the required comminution energy and semi-autogenous mill operating time to achieve the specific product particle size distribution can be estimated with high accuracy ( R2 = 0.96). Finally, the accuracy of the results was evaluated by a ball drop-weight test.

Dynamic modeling and simulation of a SAG mill for mill charge characterization

The modeling and simulation of semiautogenous (SAG) mills have been widely used in the design and optimization of mill performance in terms of its power draw, processing capacity and product size distribution. However, these models are solved under steady approximation and do not provide any information on mill charge distribution in real time. This paper attempts to characterize mill contents by solving Whiten’s first-order content-based model in the MATLAB/Simulink environment. The parameters in the model, such as breakage rate constant, discharge rate and appearance function, were estimated using process and design data collected from a gold mine operating in Alaska coupled with a nonlinear parameter estimation scheme. This model was then used to predict the dynamic responses of other key operational variables, such as mill power, bearing pressure, charge level and product size distribution. The transient response of mill behavior with respect to changes in feed size distribution, tonnage and mill feed water is also presented. This dynamic simulation approach can be used for practicing different control strategy and training purposes. The observed response of mentioned variables was validated using dynamic response data from the plant to encompass operational characteristics of SAG mill behavior.

基于混合神经遗传算法的半自磨机功率预测

There are few methods of semi-autogenous (SAG) mill power prediction in the full-scale without using long experiments. In this work, the effects of different operating parameters such as feed moisture, mass flowrate, mill load cell mass, SAG mill solid percentage, inlet and outlet water to the SAG mill and work index are studied. A total number of 185 full-scale SAG mill works are utilized to develop the artificial neural network (ANN) and the hybrid of ANN and genetic algorithm (GANN) models with relations of input and output data in the full-scale. The results show that the GANN model is more efficient than the ANN model in predicting SAG mill power. The sensitivity analysis was also performed to determine the most effective input parameters on SAG mill power. The sensitivity analysis of the GANN model shows that the work index, inlet water to the SAG mill, mill load cell weight, SAG mill solid percentage, mass flowrate and feed moisture have a direct relationship with mill power, while outlet water to the SAG mill has an inverse relationship with mill power. The results show that the GANN model could be useful to evaluate a good output to changes in input operation parameters.