Learning to Tune a Class of Controllers with Deep Reinforcement Learning

Control systems require maintenance in the form of tuning their parameters in order to maximize their performance in the face of process changes in minerals processing circuits. This work focuses on using deep reinforcement learning to train an agent to perform this maintenance continuously. A generic simulation of a first-order process with a time delay, controlled by a proportional-integral controller, was used as the training environment. Domain randomization in this environment was used to aid in generalizing the agent to unseen conditions on a physical circuit. Proximal policy optimization was used to train the agent, and hyper-parameter optimization was performed to select the optimal agent neural network size and training algorithm parameters. Two agents were tested, examining the impact of the observation space used by the agent and concluding that the best observation consists of the parameters of an auto-regressive with exogenous input model fitted to the measurements of the controlled variable. The best trained agent was deployed at an industrial comminution circuit where it was tested on two flow rate control loops. This agent improved the performance of one of these control loops but decreased the performance of the other control loop. While deep reinforcement learning does show promise in controller tuning, several challenges and directions for further study have been identified.

SIMULATION-DRIVEN DEVELOPMENT FOR COARSE COMMINUTION PROCESS - A CASE STUDY OF GEITA GOLD MINE, TANZANIA USING PLANTSMITH PROCESS SIMULATOR

A comminution process is a material size reduction and separation process which is primarily used in the aggregates and the minerals processing industry. Knowledge related to equipment’s operation, raw material properties, operational strategies, control system, maintenance, etc. is needed to design a capable plant. New needs are arising from the industry for existing operational crushing plants such as investigation for improvements, upscaling, and downscaling of the capacity. The paper presents an application of simulation-driven development for a crushing plant in an existing gold processing plant. Due to the change in ore characteristics and the need for optimizing the cost of operation, it is required to investigate the opportunities for improvement and alternative options for downscaling the capacity of the plant. A systematic process for configuring, developing, and evaluating alternative concepts using a process simulation tool is presented. The results show the process of generating knowledge for alternative crushing plant operation settings and how the choices can be selected and eliminated using boundary conditions. The evaluation presents possible improvements and alternative concepts with their opportunities and pitfalls.

Application of Cellulose Derivatives in Mineral Processing

Cellulose derivatives (CDs) have been recognized as an anionic water-soluble, non-toxic, biocompatible and biodegradable polysaccharide. The CDs have been used as a viscosity regulator, thickening agent, sizing agent and coating agent and emulsion stabilizer, electrode binder in various industries. These characteristics properties of CDs are associated with hydroxyl groups/functionalized groups present in their structure. The CDs have significant advantages in various fields including several industrial applications such as mineral processing, palletisation process, oil drilling industrial applications due to their non-toxic and selective properties. Moreover, The CDs have been extensively used as a depressant, dispersant as well as flocculants in mineral processing from various ores. During the mineral processing like flotation of sulfide minerals highly toxic inorganic species were used as dispersant and depressant which ultimately cases environmental toxicity. Therefore, there is a current need to introduce CDs as various alternative nontoxic dispersant and flocculants. This chapter emphasized an overview of the application of CDs in mineral processing including the structure, properties of the commonly used minerals processing.

Scale-Up of Decanter Centrifuges for the Particle Separation and Mechanical Dewatering in the Minerals Processing Industry by Means of a Numerical Process Model

Decanter centrifuges are frequently used for thickening, dewatering, classification, or degritting in the mining industry and various other sectors. Their use in an industrial process chain requires a sufficiently accurate prediction of the product and the machine behaviour. For this purpose, experiments on a smaller pilot-scale are carried out for scale-up of a decanter centrifuge, which is usually a major challenge. Predicting the process behaviour of decanter centrifuges from laboratory tests is rather difficult. Basically, there are two common ways of scale-up: First, via analytical methods and the law of similarity, which often requires an enormous experimental effort. Second, using numerical models, which demands a mathematically and physically precise description of the multiple processes running simultaneously in such machines. This article provides an overview of both methods for scale-up of a decanter centrifuge. The concept of a previous developed numerical approach is introduced. Pros and cons of both scale-up methods are compared and further discussed. Experiments on lab-scale, pilot-scale, and industrial-scale decanter centrifuges with two different finely dispersed calcium carbonate water suspensions were carried out and simulations were done to investigate and prove the scale-up capability and transferability of the numerical approach.

Simulating the binding of organic functional groups to aqueous calcium carbonate species

The interaction of organic molecules with mineral systems is relevant to a wide variety of scientific problems both in the environment and minerals processing. In this study, the coordination of...

Recycling Mine Tailings for a Sustainable Future Built Environment

The future sustainable built environment focuses mainly on environmental conservation and technological innovation and development. However, with infrastructure development, the consumption of raw materials such as cement, gypsum, sand, and stones increases. Therefore, use of industrial waste as raw material in construction shall be proposed as a sustainable and environment friendly alternative. Also, the higher demand for mineral commodities have led to increased mining and hence increased mining waste. The mine tailings being the wastes from rocks and minerals processing, are generally rich in Si, Ca, Al, Mg, and Fe, and also have considerable amounts of heavy metals and metalloids such as Pb, As, Co, Cu, Zn, V, and Cr. When tailings contain sulphide minerals, it may also lead to acid mine drainage. This makes the effective and efficient recycling and reuse of mine waste a major environmental concern. However, the physical, mineralogical and chemical composition of the mine tailings renders it a suitable material for use in civil engineering applications. This paper discusses the use of mine tailings of different origins for different civil engineering applications such as bricks, ceramics, fine aggregates, coarse aggregate and cementitious binders. This approach has a potential to reduce the demand on existing natural resources to face the demands of the exponentially developing infrastructure.

Nutritional diversity in accessions of African walnut (Plukenetia conophora L.) sourced from southwestern Nigeria as influenced by collection center and processing

Nut collection expeditions of Plukenetia conophora L. were made to 10 locations in southwestern Nigeria. Seeds extracted were cooked or not and were assayed for phytochemical and mineral contents. Data obtained were statistically analyzed. Omuwo-Oke accession had highest phenol and tannin whereas Ijan had highest phytate; alkaloids and glycosides were higher in Ara accession than others. Accession obtained from Aramoko gave highest iron concentration while the lowest was in Ogotun accession. The concentrations of K, Mg and Na were higher in Ogotun accession while the lowest was found in Alafe, Agbado and Aba-Oyo, respectively. Alafe accession had highest concentration of phosphorus but lower concentration was obtained in Ogotun. Zinc was highest in collection from Aramoko while Abgado accession had the lowest. Phenol, phytate, alkaloids and glycosides were higher in boiled nuts than fresh nuts but tannin was higher in fresh nuts. K, Mg, Zn and Na were higher in fresh nuts than boiled nuts. Higher concentration of Fe and P were recorded in boiled nuts. Genotype x traits biplot analysis revealed that cooked nuts from Agbado, Alafe, and Aramoko had higher values for glycosides. Alkaloid was more pronounced in fresh nut from Ara. Fresh nuts from Aba-Oyo and Agbado gave higher values for phytate. Tannin was highest in fresh nuts from Omuo-Oke and boiled nuts from Igbara-Odo while phenol was higher in boiled nut from Omuo-Oke. Biplot analysis for minerals revealed higher concentration of Zn, NFE, Na and P in boiled nuts from Aramoko, Omuo- Oke, Igbara-Odo, Ogotun and Ijan. Mg was highest in fresh nuts from Ara and Igbara-Odo. Tannin and alkaloids had negative significant relationship (r = –0.581*) but the relationship between glycosides and alkaloids was positive and significant (r = 0.520*). Zn had significant relationships with Mg and P (r = 0.590* and –0.551*, respectively). Similarly, Na had positive relationships with K and Mg (r = 0.599** and 0.491*, respectively). Mg and K had an r-value of 0.502*. Inter-correlative responses between the phytochemicals and minerals revealed that only glycosides had significant relationships with K, Mg and Zn (which were negative) as well as with P (which was positive). Data obtained suggested sufficient genetic diversity in nuts of Plukenetia sourced from southwestern Nigeria, indicating posibility for selection. Besides, cooking had signifcant (in some cases, positive) effect on the phytochemicals and minerals assayed. Key words: African walnut, accessions, phytochemical, minerals, processing

10th World Conference on Sampling and Blending (WCSB10)

The WCSB10 conference will cover the latest research and application of the Theory of Sampling (TOS) and Blending in many important technology and industry sectors: mining, exploration, minerals processing, metals refinement, cement, food & feed, agri & aqua culture, pharmaceutical production etc. WCSB10 specifically has a broader societal, industrial and environmental emphasis with a special focus on sustainable science, technology and industry. This article provides an update on WCSB10 with information on all aspects of the conference.