Uneven break optimiser
Summary of technology
In underground stoping mines, one of the most complex problems is to ensure that blasting is efficient and minimises “uneven break”. This means ensuring that as much of the mineral bearing ore is recovered (minimal underbreak or ore loss) while avoiding the inclusion of too much waste rock (minimal overbreak or dilution). This piece of highly specialised optimisation software predicts the amount of ore loss and dilution due to uneven break, and provides general recommendations to minimise uneven break prior to stope production blasting. The software uses 10 input parameters (in the areas of blasting design, geology and stope design) to predict the percentage of uneven break, and then provides recommendations on powder factor and ground support control rate values which will minimise the potential uneven break.
This technology has the potential to significantly enhance the productivity of underground stoping mines. For example, in a medium sized mine with total profit of around $20 million/year, roughly 10% of unplanned ore loss (underbreak) and dilution (overbreak) cost around $2 million/year and $10 million/year respectively. Improving this by 50% would lead to a $5-6 million increase in profit.
The software is highly innovative, using artificial neural network and fuzzy expert systems. In contrast to existing models, based on the stability graph method which predict overbreak only, it is the first practical model that predicts uneven break and is the first model that provides quantitative values to control uneven break.
Professor Erkan Topal and Dr Hyong Doo Jang are researchers at the Western Australian School of Mines within Curtin University. The research group have undertaken work in a number of aspects of optimisation of the mining process at both planning and operational levels.
Stage of development
The methodology has been developed and a stand-alone application has been coded with a user-friendly graphical interface.
Intellectual property is owned by Curtin University. There is the potential to protect novel aspects of the methodology and the software implementation ie. the source code is kept confidential.
We are looking for a supplier of software to the mining industry to collaborate on testing and improving the algorithms with a view to licensing the technology to integrate with or add to their suite of products
Roger Plumb, Business Development Manager, Commercialisation, Curtin University
Phone: +61 8 9266 4925