Blogue
Robots in the mine
Canadian mining companies hope to automate some of their ore drilling and hauling operations by 2030. (Photo : Agnico-Eagle)
Within a few years, robots will replace humans in mines, for example, autonomously drilling rock or transporting ore to the surface. How can we optimize robot work? What place should humans have in this new mining industry? These are the kinds of questions that a team from Polytechnique Montréal is addressing. Their goal and that of their partners: to automate ore extraction in the first Canadian mine by 2030.
From a Montréal command post, a worker watches what's going on in a mine in northern Québec. In that location two kilometers underground, a drill rig is working its way down a wall where a promising vein is seen. In another camera snapshot from another gallery, a shuttle loader is headed back to the surface loaded with ore. All this, taking place without a human’s physical intervention.
This surreal scene could become routine in the not too distant future – in fact, the transition has already started.
Agnico-Eagles' La Ronde mine in Abitibi is a pioneer in this field with its LTE wireless communication network. Although they’re stationed more than two kilometers underground, workers at this mine now participate in videoconference meetings to report on work progress, so there’s no more waiting for the end of the shift to come to the surface to report a snag that could disrupt the next shift. The company's LTE network also autonomously controls galleries’ ventilation, and allows operators to remotely control certain equipment. It is however a job that comes with certain limits, because dust hinders camera vision.
However, this is only a first step, and eventually, mining companies aspire to automate multiple processes. Some mining deposits could even be operated 24 hours a day, 7 days a week – all thanks to automation.
Mining data
 Professor Michel Gamache (crédit photo Jean-François Ferland)
|
Before we get to that point, there’s still a lot of work to be done. Some research groups are focusing on the robotization of machinery, while others are focusing on automating processes inside and outside mines. The latter is what interests Full Professor Michel Gamache, from Polytechnique Montréal’s Department of Mathematics and Industrial Engineering.
With his team members and his academic and industrial collaborators, Professor Gamache is interested in each link of the mineral extraction chain, from drilling to tailings management.
Keen on an example? As part of one of the projects, the group is studying various ways to exploit data generated by sensors already found on work equipment. At present, hundreds of data points are being generated without necessarily being used, information such as vibration measurements, weight, pressure, temperature, etc.
But all that may change.
"One of our objectives is to facilitate the use of this data for preventive and predictive purposes to determine the best time to perform equipment maintenance," explains Gamache. "We want to predict breakdowns, but also try to better manage preventive maintenance such as equipment oil changes or tire inflation, to synchronize operations and avoid too many production stoppages."
This analysis tool should ultimately offer diagnostics to a manager through an interface that’s the subject of another project. The group is collaborating on that project with Assistant Professor Philippe Doyon, from the Department of Mathematics and Industrial Engineering at Polytechnique Montréal.
Gamache also notes that real-time data analysis should also help limit friction points throughout the extraction process. For example, data recorded by the drill tells us a lot about the geometallurgy of the rock being extracted. "It's not just about knowing the gold content, for instance, but of understanding rock geophysics to know what impact each tonne of ore will have on the tailings pond," explains Gamache.
Increased knowledge of each load’s properties, in addition to an estimate of the weight of material transported, will permit the use of exactly the right amount of reagent to extract the ore once the load enters the processing plant. This handy trick will limit monetary costs and the environmental impact that comes with the extraction process.
"Instead of waiting for the load to come up to the surface, workers will be able to prepare for the next step," says Gamache. "The idea is also to make sure that we’re anticipating rather than reacting.
In the same vein, his group is testing the potential usefulness of digital twins in the mining sector. By reproducing the parameters of an operation in silico, the impact of certain decisions on all the mine’s activities could be tested ahead of their implementation.
destination... 2030 |
Once robots are in place in mines, their work will need to be coordinated so as to optimize ore extraction and limit the overall environmental impact of operations. (Photo : Agnico-Eagle) The various projects that Gamache's students are working on are part of MISA Group activities - a Québec-based initiative for the digital transition of the mining industry, supported by the Ministère de l’Énergie et des Ressources naturelles du Québec, and which brings together stakeholders from academia and industry. MISA's ambitious goal is to create the first autonomous mining operation by 2030, a project called Mission Mines Autonomes 2030 (MMA 2030). The approach is inspired by a Swedish initiative, wherein mining industry manufacturers have collaborated to develop autonomous vehicles adapted specifically to their industry. On the other hand, the Québec project envisages the control center of tomorrow, from which all underground mining operations will be directed, based on data analysis and artificial intelligence. "The idea is that if we can operate a mine remotely in this way, we’ll be able to operate dozens of mines from the same location," explains Professor Gamache. About 30 research projects are underway as part of this initiative, some of which are also supported by the MITACS platform. |
Comments
Commenter
* champs obligatoire