Open Cut » Geotech
The primary purpose of C29044 was to assemble a set of high‐quality data on a diverse range of characterised spoils that could be used (by C29048 Image Based Automated Characterisation of Waste Materials) to calibrate a “machine learning” model to recognise spoil category from parameters derived from simple inputs such as RGB photographs.
Eight field campaigns were carried out at seven different sites across the Sydney and Bowen Basins. The basic unit of spoil considered was a truck‐tipped load of spoil, and the study focussed on areas where spoils of different origins had been paddock tipped. Data acquired included a field categorisation according to Simmons and McManus (2004), ground‐based photographs, UAV photography, and GPS coordinates, with sampling and hyperspectral scanning of selected piles. The fieldwork was carried out over a protracted period, at the conclusion of the fieldwork, a total of 686 spoil piles had been categorised. Laboratory testing of collected samples was undertaken to confirm assessments of plasticity made in the field.
The framework for spoil categorisation and shear strength estimation, as set out in Simmons and McManus (2004) is now well established and anecdotal feedback suggests it is used successfully across the black coal mining industry in Australia. The framework, often informally referred to as the BMA framework for historical reasons, comprises two parts: one to categorise the spoil according to its physical characteristics, and the other to assign likely Mohr‐Coulomb shear strength parameters on the basis of the assigned category. Although it was not the primary purpose of project C29044, the scope of the fieldwork presented an ideal opportunity to assess and review the appropriateness of the spoil categorisation aspect of the framework, in a manner complementary to the review of shear strengths that was facilitated by project C20019.
This report presents a review and discussion of the factors that influence the shear strength of waste rock, and it considers how the categorisation attributes employed in the framework account for those factors. The appropriateness of these attributes in their use as a basis to categorise spoil was assessed against the large number of spoil categorisations that were carried out for this project.
On the basis of these considerations, the framework, its attributes and their weightings are found to be generally fit for purpose. However, the review recognised that the justifications for the framework provided by existing reference documents are limited, and written guidance required to categorise spoil in a consistent and reliable way was inadequate. It also recognised that there was confusion around the origins of the framework arising from inconsistent references to “Simmons and McManus” and “BMA”.
To provide a broader context and more detailed guidance for the application of the categorisation framework, this report includes an appendix where a complete, stand‐alone version of the spoil categorisation framework is presented. In accordance with the suggestion of the Project Coordinator and Monitors, the framework is rebranded as the “CoalSpoil Framework”, with the intention that it will be known and referred to as such in the future.
The CoalSpoil Framework document, presented in the format of a standard in Appendix A, provides:
- Clear definitions of the attributes considered to categorise spoil;
- Practical guidance to rank the attributes consistently in a field assessment;
- The process used to arrive at an integer spoil Category, based on attribute rankings; and
- A suggested correlation between Category and shear strength parameters, which explicitly incorporates the exclusions identified in ACARP project C20019.