IM March 2024 March 24 | Page 64

PRE-CONCENTRATION & ORE SORTING
recovery from waste zones as representing a profitable application assuming conveyors are long enough for the required direct gold measurement times prior to reaching the diverter system .
AI unlocking new particle sorting potential
Particle sorting has already had mentions in this article from the bulk sorting community , hence its importance across the wider preconcentration and ore sorting space .
One of the leaders in this field is TOMRA Mining , which recently invigorated its offering with the help of a sophisticated deep-learning platform called OBTAIN™ .
TOMRA has been using machine learning in its X-ray Transmission ( XRT ) and NIR sorters for the last 10 years but OBTAIN is set to open a new era in sorting , the company says .
OBTAIN goes a step further , leveraging deep learning to bring single-particle precision to high throughput particle sorting , TOMRA says .
The software uses a neuronal network to identify the properties of each particle accurately and independently of the sorter ’ s capacity , achieving new-found precision and reliability in detection and ejection . Based on its specific requirements , the mining operation has the flexibility to either enhance the throughput of the sorter while maintaining consistent sorting efficiency , or improve sorting precision – ie recovery – without compromising the existing throughput .
When asked about the maximum throughput of the COM XRT 2.0 – TOMRA ’ s biggest particle sorting machine – when equipped with OBTAIN , Stefan Jürgensen , Software Team Leader , R & D at TOMRA Mining , said the volume depended on factors like PSD and commodity-specific requirements . “ OBTAIN enhances both the throughput and the quality of the sorting process , but it does not have a fixed value for all cases ,” he told IM . “ OBTAIN ensures consistent sorting quality across different throughput levels .”
Sorting of very low-grade material often depends on sufficient liberation of the ores . Further crushing of the material increases the belt occupancy and particle count considerably . The resulting reduced distances between particles negatively impact the sorting quality in conventional sensor-based sorting . As OBTAIN brings single-particle precision to high throughput sorting , it makes the processing of low-grade ores more cost-effective than ever before , according to Jürgensen
He added : “ Traditional sensor-based sorting technologies can only maintain a certain quality level up to a limited throughput . Beyond this limit , quality declines due to constraints in
The addition of OBTAIN allowed a long-standing customer in a magnesite application to double the feed on the sorter without compromise the sorting quality . In fact , the sorting quality improved , demonstrating the efficiency and effectiveness of the OBTAIN system , Stefan Jürgensen says
physical separation , material presentation and data processing . OBTAIN enhances the detection and evaluation of each particle on the sorter , which enables higher feed rates without compromising product quality . This means that operators don ’ t have to slow down the feed to use OBTAIN .”
OBTAIN works best on ore sorting units with the new high-resolution TS100C ejection module , which is fast and extremely precise , according to Jürgensen . The module also has the added benefit of significant energy savings because it uses much less compressed air to eject the particles than other ejection systems .
On top of throughput and recovery , OBTAIN can also add value to a mining operation with a wealth of extremely detailed and accurate data , such as precise online PSD of the feed , the company says .
This online PSD analysis offers near real-time monitoring of material streams by assessing each particle , according to Jürgensen . “ It delivers continuous insights into mineral composition and crushing patterns , further enriched by sorting statistics on the detected materials ,” he said . “ This immediate , accurate observation enables operators to respond swiftly to changes , improving productivity and cost-effectiveness . It is particularly valuable in high throughput mining and mineral processing quality control , where prompt decision making is essential .”
When used in combination with TOMRA Insight , OBTAIN can provide the customer with detailed reporting on the performance of the sorter and its components to help them optimise the process , as well as enable them to plan for predictive maintenance , the company says . TOMRA partnered with two customers to test OBTAIN in real working conditions . The software has been operating for over 18 months at the Wolfram Bergbau & Hütten tungsten mine in Mittersill , Austria , where it has delivered consistent and reliable performance . The vicinity of the mine to TOMRA ’ s development team , based in Germany , made it a suitable testing ground for the first phase .
Jürgensen said Wolfram Bergbau & Hütten AG was operating its sorter at the maximum feed rate of its underground tungsten mine , with the primary objective of using OBTAIN being to improve the sorting precision . “ With the installation of OBTAIN , they not only achieved this goal but also gained the ability to sort additional , finer material ,” he said .
The second phase of testing the company has carried out to date is with a long-standing customer in a magnesite application . Jürgensen said this customer was already using its sorter at maximum capacity , but the addition of OBTAIN allowed it to double the feed on the sorter . “ Remarkably , this increase in throughput did not compromise the sorting quality ,” he said . “ In fact , the sorting quality improved , demonstrating the efficiency and effectiveness of the OBTAIN system .”
The OBTAIN software has been developed for TOMRA ’ s XRT sorters . It is available on new
SpectraFlow Analytics has recently received an order from Vale for a Crossbelt Analyzer as part of the mining company ’ s quality monitoring processes at its iron ore dispatch terminal of Ponta da Madeira , in Maranhão , Brazil .
This is the second order from Vale and the third order for a minerals Crossbelt Analyzer in South America .
The company said : “ After the very successful installation of a SpectraFlow Crossbelt Analyzer for the same application in the Terminal Ilha Guaíba , in Rio de Janeiro , this is now the second installation of a SpectraFlow Analyzer within Vale to enhance the ship loading with real-time
data .” SpectraFlow analysers are already installed in the cement , iron ore , gold , platinum , copper , potash , bauxite and recycling industries . They leverage NIR sensors , providing measurements of dry raw material composition at key points in mining , processing and refining value chains .
60 International Mining | MARCH 2024