Mining Pioneers

Inside Sandvik’s test mine in Finland

Image credit: Louise Murray

Using a combination of automation, digitalisation and electrification, an experimental mine in Finland demonstrates how the latest technologies can make the mining industry more safe, efficient and sustainable.

Mining companies rely on continuous innovation to be able to to deliver the productivity increases that drive economics in this highly cyclical sector. Electrification, automation, digitisation and data analytics are all key current trends.

E&T was given a rare opportunity to visit global engineering group Sandvik’s experimental test mine in Tampere, Finland, to view the latest in technical developments and concept vehicle testing in hard rock underground mining.

“If you can drill here you can drill anywhere,” says Jani Vilenius, director of technology development and engineering services at Sandvik Mining and Rock Solutions, referring to the granite rock at the mine. “It’s one of the ten hardest rocks in the world, making it an ideal test environment.”

A big orange mining machine underground

Image credit: Louise Murray

Established in the early 1970s, the continuously growing site contains 6km of tunnels and is used for research, development and testing of future concepts in mining technology, and for product development and prototyping. It also houses a virtual-reality simulator training suite for customers to develop skills in vehicle operations and maintenance. The test mine is also where each piece of mining machinery bought by the customer undergoes a full functionality real-world check before being shipped out.

Electrification is a huge trend in mining given that up to 60 per cent of energy costs in running an underground mine can be swallowed up in ventilation. The mine of the future is electric. New mines such as those operated by Resolute Mining in Mali, and at Newmont Goldcorp’s Borden Lake facility in Canada, are now designed to be wholly electrically powered from the outset. This minimises emissions and power consumption. The gradual replacement of diesel engines with battery-operated vehicles is already under way in older mines, with the e-option offering significant benefits over diesel-fuelled models.

Aside from the obvious removal of exhaust fumes from the underground environment and the concomitant benefits to workers, electrification also improves the mining environment in other ways, as the equipment is so much quieter – some say up to 90 per cent – than the equivalent diesel-powered machine, with an additional significant decrease in heat production.

A small underground mining truck

Image credit: Louise Murray

Battery-swapping technology avoids down time for the charging of electric machines. The instant-on torque means that the vehicles deliver up to 70 per cent faster acceleration, under load and on gradients, than their diesel-fuelled counterparts, leading to better economics during operations.

Electrification also reduces greenhouse gas emissions, fossil fuel use and improves energy efficiency. Sandvik was an early pioneer of electrical vehicles, introducing its first electric loader truck back in the 1980s, and subsequently released the first electric and automated LHD (load, haul, dump vehicle) in 2009. In over two million hours of operation, there have been zero accidents involving workers, an enviable record.

The mining industry has been crying out for a large-capacity electric truck, and in March 2022, Sandvik delivered the world’s largest battery-powered electric truck to a gold mine in Australia for further real-world testing before commercial production starts in 2023. Vilenius acknowledges: “The new concept loader has benefited from developments in the construction industry in large cities where lower CO2 emissions and noise levels at sites are even more important than in the underground environment.”

A big digging machine, yesterday

Image credit: Louise Murray

With a 65-tonne payload, the snappily named TH655B is unlikely to hold the top capacity spot for long. Impressively, the truck’s electric motors deliver 640kW of power, and it is able to haul a 65-tonne load up a 14 per cent gradient at 11.5km/h, 30 per cent faster than Sandvik’s equivalent diesel-powered vehicle. Importantly, its batteries can be swapped in less than three minutes. All this results in vital productivity increases – more tonnes moved per hour.

Many of us take for granted that at least some form of autonomous vehicles will be on our roads in the relatively near future, but spare a thought for the engineers and technologists who have had to develop autonomous systems in the dark, dirty, GPS-denied underground environment. In what can only be described as highly challenging deployment conditions, every sensor, radar, and camera system has to be proofed against the environment.

Software and AI must competently and reliably deal with the precise localisation of multiple vehicles in repetitively similar shafts, corridors and stopes where dust clouds can obfuscate the views from ordinary cameras. Critically, the whole system has also to be robust enough to facilitate remote operations, where the machinery operator may in fact be overseeing multi-vehicle operations, many kilometres away from the actual mine site.

Among the technologies being put through their paces at Tampere are scanning and mapping devices, electric and autonomous vehicles and advanced control software

Image credit: Louise Murray

With increased digitisation in a safety-critical environment comes the challenge of reliance on a network data load. The team have been moving data-intensive operations such as onboard AI physically on to the mining machines. Because of safety issues there can be zero tolerance of latency. Sandvik turned to 5G network slicing as a solution. In a world first earlier this year, Nokia in conjunction with Finnish telecom operator Telia launched its 5G Edge Slicing technology at the Tampere experimental mine (see boxout below). This is a private slice of the publicly available 4G/5G network that is reserved for the Sandvik operations and is both robust and secure. 5G Edge Slicing creates a virtual private network and is an alternative to a much more expensive private physical network.

“It’s great to be a pioneering technology leader. We have had standard 5G for two years already,” Vilenius explains. “And of course, with cyber security in mind our 5G VPN is ultra-secure. The network supports further developments in automation and digitisation in the mining environment. With this new ability to process huge amounts of data on site, we are learning how to use this capability to design better machines.”

OptiMine is Sandvik’s software suite designed for the analysis and optimisation of underground hard rock mining production – a mission control centre for mining operations. It maps and monitors the entire mining environment. Every employee and every mobile asset in the mine is tracked in real time. It is hardware-agnostic, working with any manufacturer’s machinery. The real-time mine optimisation planning and monitoring system has been in continuous development since 2004. The system is now in use in over 170 mines worldwide.

‘With this new ability to process huge amounts of data on site, we are learning how to use this capability to design better machines.’

Jani Vilenius, Sandvik

OptiMine has at its heart the production plan. Instructions to the fleet loaders and trucks are sent out and actual real-time performance data comes back in. This allows the continuous monitoring of operations and a comparison with the production plan goals. Kai Narvanen, senior automation technical expert, gives an example: “If a loader needs to shift 1,000 tonnes in an hour, but is only achieving 750 tonnes, the OptiMine user can examine the data and answer the questions – Is it the machine that is underperforming? Or has the environment changed, and the plan needs to be revised? Or is it the human operator or a vehicle supervisor shortfall that is not delivering the tonnage? Fine analysis like this was just not possible until recently.”

The data flow from multiple sources and sensors around the mine, on mining vehicles, and from post-drilling observation of the ore body, is analysed and transformed into actionable insights, including feeding into a predictive maintenance programme.

A separate software system, AutoMine, which can optionally be integrated into the OptiMine system, is the control software for multiple autonomous vehicles in a mine fleet. Obstacle detection, collision avoidance and 3D online mapping are built in.

On site in the experimental mine is a small, orange 2m x 2m autonomous drone-carrying mapping robot, the result of a partnership between Sandvik and US-based aerial robotics specialist Exyn Technologies that is the only one of its kind in the world. Once its surveying job is complete, its drone returns to the wheeled robotic base for recharging. One of its jobs as an inspection vehicle is to compare the original point cloud map generated by a 3D lidar scan – conducted either by itself or another vehicle – with the current situation. Algorithms compare the new map with the original and can also flag up areas of concern where the environment is changing due to shrinkage, bulges, shifts or slippages.

A highly automated i-series jumbo drilling machine uses iSure sensors and software to control, analyse and optimise tunnel construction and mining drill and blast processes at the rock face

Image credit: Louise Murray

Another important development task is use of data to update the production plan after a drilling operation, with a view to being able to calculate the mining operation productivity in real time by comparing the volume of orebody removed with its known percentage of ore. The autonomous inspection vehicle can launch its resident drone to survey new voids after drilling and areas where the wheeled vehicle cannot pass or areas where it is too dangerous for other forms of survey.

New lidar sensors on this and other vehicles in development have integrated processing, rather than sending their data to the cloud for processing. The separation of AI systems for autonomy and safety sub-systems is vital if humans and machines are to share the same space underground safely and intelligently.

The company has had a 5G network underground at the experimental mine in Tampere for over two years, but the new 5G Edge slice network means that safety-critical data can be processed onboard or in a local cloud. It would be too slow to upload to a remote cloud operation for processing. Onboard data can be processed in milliseconds, avoiding latency.

Another concept vehicle under test is an autonomous electric loader. It is equipped with a new suite of sensors and artificial intelligence allowing it to plan and adapt its own routes around the mine, finding the most suitable paths even in a continuously changing environment. And in keeping with the trend to remove humans from the dangerous mining environment as far as possible, the loader has no cabin and is a battery-driven driveline machine.

Productivity gains have generally been exceeding expectations as the various technologies developed in Tampere move into the real world. Miners have been dreaming about access to these technologies for years, and it is clear that as the industry moves into working hard-rock underground mines around the world productivity and human safety are improving. And not to be entirely flippant, what’s not to like about huge orange robots? 


5G Edge Slicing technology

Edge slicing creates a secure, dynamically reserved portion of a 4G/5G bandwidth reserved for private use – a 4G/5G virtual private network. Essentially it can bring extremely fast and localised cloud processing to remote areas or indeed any business in need of a secure local network. In a world first, this was installed by Nokia and network partner Telia at the Sandvik experimental mine in Tampere in the spring of 2022.

“Before we installed the edge slice, packets of data from a user in Tampere would route through to Helsinki, some 200km away, and back again, with concomitant latency and a data processing delay,” explains Nokia’s head of network slicing, Mika Uusitalo. “Even worse, if that data was being processed in the nearest Amazon cloud data centre, it would be going from Tampere to Stockholm and back, over 700km each way. While the average customer would not even notice this, this is just not a functional service where the safety and control of large complex sites, multiple autonomous vehicles and human workers is concerned.”

Uusitalo adds that the system is dynamic and can be boosted for business-critical services on demand. Equally, unused portions of the slice can be returned to the public network when there is less activity, such as over the weekend.

There are no new basestations needed, just new software upgrades. The whole management of the edge slice end-to-end is deployed at the telecom operator’s existing network operations centre. The cloud service provider can bring their applications into the local edge point of presence, or the business premises.

Outside the mining environment, 5G slicing technology is also being tested in several business case studies – factories where video analytics, remote-controlled robots and telemetry combine and in digital plant monitoring, control and management.

This article was updated on 20 October 2022 to clarify that Exyn Technologies' drones are being used by Sandvik as part of a partnership and not a corporate acquisition.

Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.

Recent articles