Tag Archives: 3D rugged machine vision

MotionMetrics-BeltMetrics

Weir eyes game-changing energy intensity reductions with ore characterisation project

With an extensive footprint from the rock face all the way through to tailings, it was only a matter of time before the Weir Group decided to enter the ore sorting game.

In recent years, the company has re-focused as a pureplay mining and aggregates company that can provide value throughout the flowsheet.

The company ditched its oil & gas exposure and added to its process plant and tailings remit with the acquisition of ESCO, a front-end-focused mining technology company with leading market share in the ground engaging tool (GET) segment.

Having more recently incorporated Motion Metrics into the mix – now within the ESCO division – it is embarking on a project that could have positive ramifications throughout the wider Weir Group offering.

Motion Metrics is a developer of artificial intelligence (AI) and 3D rugged machine vision technology. Its smart, rugged cameras monitor and provide data on equipment performance, faults, payloads and rock fragmentation (read: particle size distribution (PSD)). This data is then analysed using embedded and cloud-based AI to provide real-time feedback to the mining operation.

Initially developed for GET applications, these technologies have recently been extended into a suite of products and solutions that can be applied from drill and blast through to primary processing. Motion Metrics has, in the process, built up an impressive customer base and income stream, performing well since the acquisition.

Weir has outlined a £500 million-plus ($604 million-plus) emerging digital market opportunity for the entity, with much of this hinged on rugged machine vision technology, its sophisticated digital platform and the ability to add ore sensing to its offering.

This became clear at the company’s recent Capital Markets Day during a presentation from Chris Carpenter, VP of Technology for Weir ESCO.

Sensing, not sorting

At this event, Carpenter said the company was combining Motion Metrics’ PSD capability with ore characterisation technology to explore high-value opportunities for its clients.

“Looking further out, we believe ore characterisation…has the potential to transform mining by moving less rock, using less energy and creating less waste,” he said during his presentation. “Ore characterisation technology, which is underpinned by sophisticated sensing systems, captures critical data on properties and composition of rock, including rock hardness and mineral and moisture content.”

“When coupled with Motion Metrics fragmentation analysis technology, it has the potential to be a game changer, giving miners a full picture of the size and characteristics of rocks.”

Weir has outlined a £500 million-plus emerging digital market opportunity for Motion Metrics, with much of this hinged on rugged machine vision technology, its sophisticated digital platform and the ability to add ore sensing to its offering

This concept is not new. Measuring the quality of ore has been spoken of for decades and, more recently, has become a reality with the likes of MineSense, NextOre, IMA Engineering, Scantech, Malvern Panalytical and Rados International, among others, all having trialled technology or deployed commercial solutions across multiple commodities and sites.

Metso Outotec, one of Weir Minerals’ big competitors in the plant and tailings arena, has also spoken of the potential for bulk ore sorting by using its existing portfolio of material handling modules, crushing stations, mobile crushing equipment and bulk material handling solutions as the basis, while incorporating sensors from other vendors.

Weir believes it is one step ahead of its OEM counterpart in its pursuit of ore sorting, even if Carpenter is only referring to the trials currently being conducted at an unnamed copper mine as “ore characterisation” studies.

“With the acquisition of Motion Metrics, what we essentially bought was the ruggedised vision systems used in both mobile and fixed applications,” he told IM in January. “While the ore sensing piece is by no means trivial, the integrated AI capabilities and digital infrastructure that allows the data to be transported via a variety of avenues is incredibly important.

“Being able to pick up the data is one thing but being able to transport that data to the right people in a secure, accurate and timely manner is something different altogether.”

With a portfolio that includes LoaderMetrics™, BeltMetrics™, TruckMetrics™ and CrusherMetrics™, Motion Metrics and the Weir ESCO R&D team had several potential applications to start its ore characterisation journey with.

The company has settled on a BeltMetrics installation for its first trial, with Carpenter confirming the sensing solution under the microscope is currently positioned above a conveyor that is directly after the crusher in the flowsheet.

“We feel we will learn quickest over a conveyor belt, so it is really an expansion of the existing BeltMetrics solution that we will start with,” he said.

The sensing options open to Motion Metrics for this trial were also vast, with the aforementioned ore sorting vendors using the likes of X-ray Fluorescence, magnetic resonance, prompt gamma neutron activation analysis, pulsed fast thermal neutron activation, and others within their solutions.

Motion Metrics has chosen to incorporate hyperspectral imaging into its PSD mix.

Carpenter explained: “When you think about ore characterisation, we are just moving from a visual spectrum base with Motion Metrics vision-based systems to the expanded light spectrum for gathering data and making decisions. This is all being built on the established digital platform the company has.”

The company is not alone in using this type of technology. MineSense has spoken of trials using multispectral sensing technologies, while Australia-based Plotlogic has been tapping hyperspectral imaging to provide precision orebody knowledge prior to mining.

Collaborating on energy intensity reductions

Safety, scalability and flexibility were three factors taken into account with the hyperspectral imaging decision, but Carpenter was also aware of the potential limitations in using such technology.

Mines will need to be willing to make some changes and invest in alternative infrastructure to leverage the most value out of the solution the company is putting forward.

“That is where productivity partnerships that we spoke about on the Capital Markets Day are going to be really important,” he said. “It is going to be essential to collaborate with customers.”

The initial collaboration with the trial mine site looks to be extensive, stretching from the back end of December throughout 2023.

The site is already equipped with a significant amount of Weir Minerals and ESCO equipment, so the collaboration appears to have started well before this trial.

“Throughout the year, we will have the opportunity to make enhancements; starting out with an initial system that is upgraded,” Carpenter said. “By the end of the year, we should have high confidence of having something ready to commercialise. It could also be that we have other trials running concurrently with this one to extend the learnings.”

The two primary key performance indicators for the trial surround accuracy and speed, with Carpenter saying the company is targeting to at least meet the metrics competing technologies have been promoting over recent years.

“In both cases, we are well equipped to measure both and – in the initial phase – we are performing well,” Carpenter said.

“Right now, when they (the mine site) carry out an assay, they have to stop the conveyor belt, take a sample off and send it to a lab. At best, the feedback takes hours, if not days. Motion Metrics has done a really good job of building the sensors, algorithms and platforms to process the data coming from above that belt very quickly.”

There are a team of very experienced, PhD-equipped personnel currently working on this trial, monitoring the real-time results from Motion Metrics’ base in Vancouver, however there is a Weir network across the globe watching and waiting for news.

A sensor above a conveyor belt able to provide ore characterisation data is step one. Step two will most likely involve leveraging this data to provide insights as well as initiate downstream actions.

Then, there is the potential to equip these sensors for the pit on an excavator or wheel loader – which introduces many additional challenges both Motion Metrics and ESCO are aware of. Understanding exactly what is in the bank or going in the bucket will be critical to improving operational efficiencies.

These are longer-term goals that Motion Metrics, ESCO, Weir Minerals and Carpenter are cognisant of – and excited about – that may provide the true value to customers throughout the flowsheet.

“What is exciting for us is that – as may be obvious – the further upstream you can make some good decisions, the more energy you can save downstream,” he said. “As you get into some of the other processing elements in the plant, there are sustainability benefits to be had – a more efficient use of reagents to liberate the elements, a more efficient grinding setup based on ore characteristics, a reduction in water use, etc.

“The driver for this has really been sustainability and energy reduction. It is all about reducing the energy intensity associated with ore.

“We feel we are well equipped and in a good position to deliver on this and provide the industry with the step change in sustainability that it requires.”