In partnership with engineering firm AMOG Consulting, remote monitoring specialist Omniflex is developing an oversize detection system for the global mining industry. The system uses sensors on the tray and machine-learning techniques to detect the size of ore rubble as it is loaded into haul trucks to be carried from the blast site to the crushing plant. This information is wirelessly communicated through Omniflex’s gateway in the cab of the truck to the driver and to central data storage for later analysis.
Oversize ore is a major hurdle faced by mining industries worldwide. Mines often operate by blasting mine and pit faces with explosives, which leaves rubble in all shapes and sizes, from small pebbles to metre-wide boulders. This rubble is then loaded onto large haul trucks and transported to an ore crusher.
Limitations at the crusher
Ore crushers have a limit to the size of rocks they can crush, and all too often oversize rubble reaches the crusher. This can block and damage the crusher, meaning production is often halted while blockages are removed, or parts are replaced. Oversize ore also causes huge losses in operational productivity while material is retrieved from the jammed crusher and sent on detours to secondary breaking areas. Furthermore, many mines simply discard oversize pieces of ore when secondary breaking processes are not economically viable, making for clear drops in operational efficiency.
“Oversize rubble can lead to decreases in productivity up to 20 percent,” said David Celine, managing director at Omniflex Ltd. “The costs associated with this are staggering, so real-time oversize detection is an area that has been in dire need of technological innovation. Together, Omniflex and AMOG are creating the ideal solution.”
The sensors, designed by AMOG and implemented by Omniflex, detect signals onboard the truck as ore is dumped into the tray. The data is processed using machine-learning techniques to estimate the size of the rocks as they are dumped into the truck. This information is then communicated wirelessly through Omniflex’s gateway system to the driver to notify them of the oversize load before they reach the crusher. The technology has demonstrated a detection accuracy of above 80%, which is a vast improvement over the current detection methods that are estimated to be less than 5% accurate.
“In this way, oversize rocks should never reach the crusher. When these systems are thoroughly integrated into mining operations, the entire problem of oversize ore can be avoided with minor effort from engineers and little additional training for operators,” continued Celine. “The fully wireless and web-based monitoring system makes for unparalleled flexibility, modularity and scalability, meaning that equipment can be freely added or removed from the monitoring system as the mining operation expands and shrinks in response to international market shifts.
Platform for other systems
“The flexibility of the web-based system makes it a great foundation for other monitoring activities around the mine. Oversize management is just one variable that site managers and engineers are interested in keeping track of for Load and Haulage. Road condition, driver safety and equipment condition are just a few other examples that have easily and quickly been appended to the system.”
This project has been part-funded by the METS Ignited Collaborative Funds, part of the Australian Government’s Industry Growth Centres Initiative, which has kickstarted many innovative mining equipment, technology and services (METS) projects over recent years.
METS Ignited focuses on projects with commercial viability, and while this project focuses on oversize detection specifically, the sensing and analysis technology has seen interest from other sectors, such as marine, ports, road and rail transportation.
| Tel: | +27 31 207 7466 |
| Email: | sales@omniflex.com |
| www: | www.omniflex.com |
| Articles: | More information and articles about Omniflex Remote Monitoring Specialists |
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version