Scientists from the Perm National Research Polytechnic University (PNIPU) have developed a simple method for controlling foam flotation in the production of potash fertilizer, which makes it possible to automate and at the same time increase the efficiency of the process, the press service of the university. April 25th.
One of the most important components of fertilizers is potassium chloride (KCl). It is in great demand among agricultural producers around the world. Other industries where potassium chloride is widely used are the chemical and pharmaceutical industries.
This potassium salt is extracted from natural minerals containing potassium salts, such as sylvinite and carnallite, by extracting it from the crushed ore by the foam flotation method in special flotation machines. The control of these machines is carried out by a flotation technologist.
Manually adjusts the feed of components based on the appearance of the foam, to which KCl or gangue particles adhere, depending on the type of flotation machine. However, the float is usually one for several machines in the flotation department, which reduces the efficiency of its operations.
PNIPU specialists decided to automate the process of recognizing the state of the foam, eliminate the influence of the human factor, increase the extraction of the necessary components from the mineral and at the same time reduce the environmental load on the environment.
The authors presented the results of the study of the new method in the article “Development and adjustment of an algorithm for software identification of deviations in the foam layer of a potassium flotation machine”, published in the journal “SUSU Bulletin” . Series “Computer technologies, control, radio electronics” volume 23, no. 1 for 2023.
The flotation process involves supplying air through a slurry (a suspension of finely ground ore in water with reagents), which creates bubbles in which waste rock particles float during suspension flotation (removal of insoluble waste) and during flotation Sylvin, a useful product. , in this case potassium chloride.
The foam is swept into the receiving tray by a rotating foam cannon. Since the composition of the mineral can be heterogeneous, its change can lead to incomplete flotation and a significant part of potassium chloride will go to landfills. In this case, visually such a phenomenon will be seen as a change in the appearance of the foam, which floaters in this case call “bald”, “uneven” or “not red”.
These visual deviations from the normal flotation process can be recognized by a computer vision system and send a control signal to adjust process parameters faster than the operator.
These systems have already been used successfully in the flotation of polymetallic minerals. But there the foam bubbles are contrasting, with almost black edges, and in the sylvite flotation machine the foam is discolored, making it difficult to recognize the edges of the bubbles to evaluate their size.
PNIPU scientists proposed their own system to recognize foam quality: instead of evaluating bubbles by determining their edges, consider the glow of their surface from a point light source.
Additionally, anti-glare spots (black dots in bubbles with a distorted shape) are also evaluated using data from a video camera. Correct adjustment of image preprocessing algorithms allows foam parameters to be evaluated quickly and reliably.
The head of the department of “Automation of Technological Processes” of the Berezniki branch of the PNRPU, Professor and Doctor of Technical Sciences Andrei Zatonsky, said that they tested their new program to evaluate the flotation process on video streams obtained on different machines of PJSC Uralkali and LLC Eurochem – Usolsky Potash Plant.
After that, the developers improved their algorithm by improving anti-glare accounting and statistical processing of glare characteristics for earlier recognition of transient floating processes.
“Our method has demonstrated great precision: at some points the error value changes between 0.5% and 1.5%, but in most cases the results are completely consistent”said the scientist.
Zatonsky noted that one statistical characteristic of glare well describes the beginning of the transition process, and the other – its end, which was taken into account by the developers of the algorithm when building a system for regulating or signaling deviations.
The development of PNRPU scientists will make it possible to increase the efficiency of potash ore enrichment and automate the processes of obtaining components for fertilizers. The new glare method was also suitable for recognizing the size of mining granules (lumps of ore concentrate) and sheet defects, its developers note.
Source: Rossa Primavera
I am Michael Melvin, an experienced news writer with a passion for uncovering stories and bringing them to the public. I have been working in the news industry for over five years now, and my work has been published on multiple websites. As an author at 24 News Reporters, I cover world section of current events stories that are both informative and captivating to read.