Orebody characterisation using machine learning and MWD data

Описание: Geotechnical engineer and Curtin PhD student Daniel Goldstein has developed a technology that rapidly analyses mining drillhole data and generates orebody knowledge insights through high-resolution models. Due to the expense of exploration drilling, inadequate geotechnical characterisation of rock masses before open-pit mining may cause inaccurate predictions that can lead to mine site instability, posing risks to safety and mine site viability.  

Mr Goldstein has developed an algorithm to analyse data from a current drilling output recording method, Measure-While-Drilling (or MWD), and provide better forecasts of geotechnical conditions and identify geotechnical hazards prior to mining.  
The technology will help mining engineers, geotechnical engineers and geologists in all surface mining companies where exploration and mine production drilling is employed.  The system can also utilise pre-existing datasets. The technology is currently at the stage of simulation. 

The project has been recognised at the 2022 Curtinnovation Awards as the winner of the Student Prize.

Team: Mr Daniel Goldstein.
Support: Curtin University.

Learn more: http://curtin.edu/curtinnovationawards