Machine learning is a subcategory of artificial intelligence, which aims to make computers capable of solving complex problems without being explicitly programmed. Availability of large datasets, development of effective algorithms, and access to the powerful computers have resulted in the unprecedented success of machine learning in recent years. This powerful tool has been employed in a plethora of science and engineering domains including mining and minerals industry. Considering the ever-increasing global demand for raw materials, complexities of the geological structure of ore deposits, and decreasing ore grade, high-quality and extensive mineralogical information is required. Comprehensive analyses of such invaluable information call for advanced and powerful techniques including machine learning. This paper presents a systematic review of the efforts that have been dedicated to the development of machine learning-based solutions for better utilizing mineralogical data in mining and mineral studies. To that end, we investigate the main reasons behind the superiority of machine learning in the relevant literature, machine learning algorithms that have been deployed, input data, concerned outputs, as well as the general trends in the subject area.

The technologies used in mineral process engineering are evolving. The digital mineral processing solutions are based on advances in our ability to instrumentally measure phenomena at several stages of the beneficiation circuit, manage the data in real-time, and to analyze these data using machine learning to develop the next generation of process control. The main purpose of this study is to overview various digital solutions for mineral processing plants and characterization laboratories while emphasizing their utilization in the current state of the digitization process of the GTK Mintec. This study highlights the specialized digital technologies that are particularly relevant for mineral processing and beneficiation. The digital solutions studied in this article include digital twin, machine vision, information management system, sensors, smart equipment, machine learning techniques, process control system, robotic cell, and Internet of Things applied across the whole chain of studying materials from the mineralogical examinations through the bench-scale studies to the pilot test trials. The aim is to provide a clear view on the different aspects of digitizing mineral processing plants based upon the lessons learned from the development plans in GTK Mintec.

The article presents the results of investigations on three lithological types of Polish copper ore: sandstone ore, carbonate ore, and shale ore. According to X-ray diffraction analysis, sandstone samples can be classified as sandstone with dolomite binder and partly clay binder; shale—as dolomitic slate with a high proportion of clay with elevated organic matter content; while dolomite has a high organic content. Five particle-sized fractions (16–18 mm, 18–20 mm, 20–25 mm, 25–31.5 mm, and 31.5–45 mm) of each lithological type were prepared. A single-axis slow-compression test was performed on single particles to determine the value of the crushing force. The Weibull distribution was used to approximate the strength distribution models and cumulative strength distribution functions for each of the materials. The residual deviation and non-linear correlation coefficient were calculated in order to assess the fitting of the model function to empirical data. In addition, the impact of particle size on the strength of the raw material was separately investigated for the hard (dolomite and shale) and soft brittle material (sandstone).

The aim of the paper is an assessment of geometrical properties of regular and irregular particles of chalcedonite enrichment products carried out in a laboratory ring jig. The investigative program included experiments of aggregate enrichment, along with visual analyses made for the obtained products, separately for regular and irregular particles. Several shape coefficients were calculated, and the most effective ones in terms of assessment of particle regularity were selected from among them. Particle size distributions for feed and enrichment products were also determined using the idea of minimum Feret’s diameter, and the intensity of dust emission by individual products was measured as well. The results obtained by the visual system were discussed in the context of their application in the assessment of enrichment operations carried out in a water jig.

The influence of the physical, geometric and chemical properties of particles on the results of aggregate separation by means of a laboratory ring jig is presented in this paper. The experiment was based on separation of chalcedonite particles in a narrow particle size fraction composed separately of regular and irregular particles, which was prepared in accordance with patent inventions. On its basis, the geometric properties—projection diameter and (volumetric and dynamic) shape coefficients—as well as physical properties—particle density—were determined in products of the regular and irregular particles. The terminal settling velocities of the regular and irregular particles were calculated for a randomly selected sample of particles in each obtained separation product. The statistical analysis of the geometric properties of the particles allowed to evaluate the influence of these parameters on aggregate processing with respect to selection of particles homogenous in terms of their shapes. The comparison of the particle shapes’ influence on the chalcedonite feed separation effects was made by the means of the values of the shape coefficients: the dynamic and volumetric ones. Additionally, tests were carried out using Raman spectroscopy in order to determine the mechanisms of density change in the aggregate. The research goal was realised through detecting and analysing the polymorphic forms of the silica and allogenic minerals precipitated on the surface and inside the chalcedonite particles.

The paper presents the results of analyze the particle size distribution function of comminution products of dolomitic type of copper ore. The breakage tests for single irregular particles were performed with using a hydraulic press device. The authors prepared five particle size fractions of each material, within ranges: 16-18 mm, 18-20 mm, 20-25 mm, 25-31,5 mm and 31-45 mm. The particle size distribution function of single-particle breakage test was calculated separately for each size fraction. In addition, the cumulative particle size distribution function for five particle size fractions was presented. In theoretical part the study of applied functions of particle size distribution for comminution a set of particles and models of crushing single particles was performed. In that paper the curves of the particle size distribution were approximated by the three-parameter function, which parameters depend on the particle strength and material type. For conformity assessment the model distribution function to the empirical distribution function a residual deviation and non-linear correlation coefficient were calculated. The three-parameter function approximating agrees well with the particle size distribution obtained from experimental data. The dependence of the parameters of a particle size distribution function on the dolomite particle strength was presented. The results indicate the identity of single particle grinding mechanism by slow compression of irregular particles of dolomitic type of copper ore, regardless of the initial particle size.