Coal Grain Analysis: Managing, Visualising, and Understanding Data for Coal Characterisation

Mr Paul Mcphee¹

¹CSIRO, Pullenvale, Australia

A commerically available image based coal characterisation system has been implemented to improve the characterisation of coal samples. Key improvements over previous systems include the characterisation of individual grains, use of very large colour images containing large numbers of particles to improve classification statistics and more automated processing.

Characterisation of coal is fundamental to understanding the chemical and physical behaviour of the coal and allows the evaluation of exploration samples, optimising the recovery of fine coal via flotation processes and improving coal utilisation for coke making or power generation.

Characterisation requires identification of the different components in the coal grains. Due to the mixing of blends and difference in cutting angles during sample preparation, there is a significant benefit in estimating the distribution on a per grain basis.

We use a non-linear least square algorithm (Levenberg–Marquardt) to robustly fit the underlying distribution components on each grain and use the data obtained to characterise it.

The software we developed allows the automatic analysis of large coal images which provide reliable statistics on the distribution of coal types and impurities.

Unique features of our software include:

– Single grain characterisation

– Enhanced particle separation

– Support for large images

Future work includes:

– Blend partitioning

– Handling of larger image files up to 40GB (tile based processing)

– Further automation with the help of machine learning algorithms

Biography:

Paul McPhee is a recent Computer Science graduate from the University of Queensland. Paul began working at CSIRO 18 months ago and has since contributed to research in Mining Automation and Coal Logistics. Paul is the main support engineer for CSIRO’s commercially available Coal Grain Analysis software tool and also contributes to its ongoing development. Prior to joining CSIRO Paul worked as a software developer in the entertainment industry.