Using Cell Network Technology to Monitor Mine Waste
The goal is not just to predict, but to solve the problem. – Wenying Liu
It’s hard to wrap one’s head around the sheer volume of ore and waste rock produced by mineral mining – 37.6 billion metric tons, or roughly 6,200 Great Pyramids, globally, every year. The great environmental challenge of the industry is to reliably predict the quantity and quality of drainage generated from waste rock, and mitigate the damage from contaminants leaching into the surrounding environment.
Researchers from UBC’s Bradshaw Research Institute for Minerals and Mining (BRIMM) have partnered with Rogers Communications (Rogers), Mitacs, and the UBC Data Science Institute in a new, high-tech approach to measuring and processing this mining waste. Recent advances in cell network technology and artificial intelligence have made it possible to monitor key physical and chemical properties inside a waste rock pile- in real time, offering an opportunity to manage disposal procedures and dramatically reduce the long-term environmental risks.
Waste rock piles are highly heterogeneous, each with a unique, complex mix of properties affected by factors such as climate, heat transport, geochemical reactions, and microbial activities, making them a challenge to properly manage. “We’re trying to understand and predict the release of various contaminants from mining waste rock using the 5G-enabled sensor networks to monitor the key parameters,” says Wenying Liu, associate professor in the Department of Materials Engineering, and co-investigator for the project. “Then we’ll use machine learning to analyze the data to predict what’s going to happen with these contaminants, which will allow us to guide management of the waste rock pile and reduce pollution. Machine learning is primarily algorithm-driven, and capable of analyzing these large, varied datasets like these to make more accurate predictions.”
The integration of 5G technology is a giant leap forward for data collection in a mining environment where traditional monitoring technology – such as radio frequencies and physical cables – severely limits the available bandwidth, transmission rate, and volume of data. “The bandwidth that 5G offers is crucial for handling the large amount of data from the numerous sensors required in this kind of environmental monitoring,” notes Roger Beckie, Professor in UBC’s Department of Earth, Ocean, and Atmospheric Sciences, and the study’s other lead investigator.
With their first year of funding in place, the team is building a sample waste rock pile in a laboratory to set up and calibrate the 5G sensor network under controlled conditions. “We don’t want to go out into the field right away, not until we work out all the kinks in a nice, controlled setting,” adds Roger Beckie. “We can control how much sulfate goes in, and we can say this is what the sensor should read because we know what we put into it.”
Using the data collected by the sensor network, they will develop the machine learning algorithms to process and analyze the collected information about what is going into and coming out of the waste rock – a complicated and ever-changing mix of temperature, moisture, pH level, gas composition, drainage rate, chemistry of outflow, and much more – continuously calibrating the algorithm until they have a working model they can take into the field and test on a live mine site.
Ultimately, mine operators can use the sensor network and computer model to guide their decisions on how to manage the waste material. “Decision makers will now have this data coming out of our model to answer questions like, Do we need to treat the water? How much? To what extent? And what is the problematic pollutant?” says Liu. “The eventual goal is not just to predict, but to solve the problem.”
“Most of these piles are going to be actively managed,” adds Beckie. “So that’s probably going to be one of the biggest operational decisions – where you pile what kind of material. And then they can monitor the drainage quality and quantity, and adjust their treatment capacity to manage contaminated loads.”
The process from lab setup to live case studies will take about three years, allowing the team to rigorously test and refine their methodology, ensuring that the system they develop is robust enough for the complex and diverse environments of active mine sites.
The novel approach, combining environmental monitoring with predictive analytics, comes at a crucial time for the mining industry, as the increasing frequency of extreme weather events due to climate change poses new risks to mine waste management that require a more proactive response.
More than just a monitoring system, the team is building a new scientific approach to mine waste management – a complete system that monitors waste rock, measures inputs and outputs, processes the data into useful information, and suggests a course of action to mitigate leaching. By integrating 5G technology and machine learning, the team is not only addressing current environmental challenges, but paving the way for a more sustainable future in the mining industry, potentially revolutionizing the approach to mine waste management.
