AI is greatly benefitting comminution and production

Artificial intelligence (AI) is transforming the mining value chain from production to comminution and beyond. At the production end of the value chain is PETRA Data Science’s AI prototype, ShovelVision, which took out the Komatsu Transform Mining even last month.

The prototype was the team’s response to the ground engaging tools challenge – detect and track missing bucket teeth.

The other three challenges set by Komatsu were:

  • Underground communications and localisation: Get location and machine data
  • Real-time material characterisation
  • Efficient ore extraction: Ore definition and low impact extraction.

PETRA Data Science CEO Dr Penny Stewart said she was excited by the win. “Thank you Unearthed, Komatsu and mentors for this brilliant opportunity. Thanks Simon Cohn from MEC Mining for sharing his strategies for success in the mining sector,” she said.

This application of AI builds on PETRA Data Science’s experience applying its expertise in comminution, such as the deployment of SAG mill overload downtime prevention algorithms at Newcrest’s Lihir mine in PNG.

“Since then, we haven’t had any overloads and, typically, we have 100 overloads per year,” said Newcrest CFO Gerard Bond.

The benefits of AI in comminution was also highlighted in a recent article by Randolph Pax in the AusIMM Bulletin. Pax, Principal at RAP Innovation and Development, examined the development of the non-contact acoustic array approach to diagnosing, controlling and understanding the internal processes of SAG and AG mills.
He highlighted how using a microphone array determined the behaviour of the particles inside a tumbling mill by analysing “the dynamics of the signals generated by the diversity of physical interactions that these particles undergo”.

Pax compares the machine acoustic measurement approach with listening to music and discerning the sounds from the individual instruments.

“The fast (80 ms) measurement has provided new insight of internal processes of a mill. The acoustic determinations demonstrate the sensitivity and value of using microphones for real-time measurement,” he said.

Pax presents a MMG Century mine case study of the technology in action.

PETRA Data Science CEO Dr Penny Stewart (centre) with her team and Komatsu representatives (Photo courtesy of Unearthed).