DYNOSORT aims to optimize raw material recovery in a pilot plant by applying a novel simulation-based approach that allows the selection of the optimal sensor for sensor-based sorting of raw materials. The set objectives are: 1.) to integrate machine-learning into the simulation-based approach 2.) to reduce analytical costs (in comparison to the previous case study) in order to offer an attractive service to the industry, and 3.) to extend the DIAMO’s current sorting process by further separating valuable components into pre-concentrates for further beneficiation.
In the final project year, the consortium will bid for an upscaling project to further validate scalability of the approach and create a spin-off offering optimized ore sorting solutions to the mining industry.
The future upscaling project will carry out an unprecedented case-study that gathers widespread interest in the raw materials sector – the reprocessing of several dozens of tons of polymetallic ore from stockpiles in Czech Republic. Optimized sensors and sorting routines will separate the starting material into a minimum of four different products with elevated compositions of different elements. The products will qualify as marketable pre-concentrates that can be further beneficiated.
Letters of Intent (LOIs) from several companies show that a need for sensor optimization in sorting of (secondary or primary) raw materials exists. Cost- and time-efficient optimization services that provide all benefits from time-taking mineralogical analysis and simulation-based models represent the greatest market potential in the manufacturing of sorting machines. The state-of-the-art does not represent these services. Approaches that would fulfill these needs are exclusively available on a high-professional and scientific level. Instead, suppliers of sorting machines follow a business trend that offers services to individually adjust and optimize sorting machines on the basis of empirical tests. Customers from the mining and recycling industry commonly use sorting test-sites in order to determine an ideal sensor-setting. As the installation of different sensor-settings is expansive and time taking, these approaches will never foresee the full potential of ideal sorting with respect to a high number of suitable sensors. DYNOSORT is capable of testing a wide range of suitable sensors numerically and the project consortium is aware of its outstanding potential. The project consortium wants to professionalize the state-of-the-art of optimizing sensor-based sorting in the raw materials sector. Therefore, DYNOSORT has the potential to supersede the empirical sensor selection.
The outcome would be a tremendous potential for economic and environmental benefits in industrial process streams due to: higher quality of pre-concentrates, reduced energy consumption in subsequent processing steps, and reduced transportation of host rock material. This outcome meets the KIC target to gain 20 cases that induce 15% savings due to higher material and energy efficiency. Other KIC targets are met by:
- the project’s co-funding that represents industrial investment into a new pilot plant
- technology is tested in the RIS country
- the use of formerly unused deposits to recover CRMs and raw materials
- the progress of a new, sustainable, and best available technology