Published in Australian Longwall Magazine

The computer simulation method realistically simulates the roof behaviour, rock bolt forces and stress redistribution around a mine’s roadway – an important achievement for Emerald in managing severe horizontal stress.

The collaborative project was undertaken a year ago by United States safety body the National Institute for Occupational Safety and Health (NIOSH), RAG Pennsylvania and SCT Operations.

SCT Operations managing director Winton Gale said the approach differed from traditional methods and was developed from observing and monitoring rock fracture in mining operations.

“Through an emphasis on detail we are able to make sure we match the model to the mechanics of what’s going on in the rock,” Gale said. “We tried to fit the model to nature rather than trying to get nature to fit a pre-existing model.”

Gale emphasised that given how complex failure mechanisms are, computer simulation required a detailed geotechnical characterisation of the strata. Key factors to be quantified were the rock intact and post failure strengths, stiffness, in situ stresses, permeability and bedding plane characteristics.

The model should be able to simulate a wide range of interacting and complex failure modes in a manner independent from input generalisations. Generalisations about rock properties had been found to limit the model’s capability to reproduce actual rock mass behaviour, Gale said.

For example, at Emerald the rock properties input into the model were derived from extensive laboratory testing, allowing much greater and more realistic geological detail. These studies allowed the element size to be reduced from ‘half a house’ to about the size of a brick, fine enough to capture geologic variations that may be important at the coal mine entry scale.

The tailgate of the 11 North longwall at Emerald was chosen for the study because severe loading to the crosscut and tailgate entry after start-up was anticipated.

“The location thus provided a unique opportunity to study the roof failure process as the applied horizontal stress increased during the progression from development through longwall mining,” Gale said.

Once the model was developed it was compared with actual field data to verify the computational method applied.

This comparison between the model results and the field measurements indicated the model effectively simulated the critical elements of the actual roadway’s behaviour. Using roof extensometer data it was found the roof deformation pattern was captured in the model and roof bolt forces were realistically simulated.

An interesting observation was that cable bolts initially have little effect on either the roof deformation or the roof bolt loads. Cable bolts appear to develop full capacity after the roof bolts yield at higher levels of roof stress.

These comparisons indicated the model’s results were consistent with the monitored and observed behaviour of the roadway, which means the mine will now be able to make a more realistic assessment of various bolt patterns within the anticipated stress conditions.

With confidence gained, the model was used as a baseline for additional simulations that evaluated the expected performance of alternative roof support systems.

Beyond the direct benefits to Emerald, the study also established benchmark data for future applications of numerical modelling to US underground coal mining.

It also evaluated the performance of roof bolts in typical US geological conditions, investigated the interaction between roof and cable bolts, and explored how well numerical modelling could aid in analysing complex ground control problems.

Gale said results of the study demonstrated that sophisticated numerical models could obtain very realistic results, so long as the details of the geology and the rock properties (both pre- and post-failure) were replicated on a very fine scale within the model and the model could simulate the broad and complex range of failure modes which occurred underground.

“Modelling to assist with mine planning and support design can be done on a comparative basis to assess the performance of different support options. If ground characterisation is satisfactorily achieved, then the results can be extended to assess ground behaviour in absolute terms. Dedicated geotechnical monitoring and strata characterisation is necessary to fully utilise, and extend, the benefits which can be obtained,” he said.

This article is based on a paper presented by Winton Gale at the 23rd International Conference on Ground Control in Mining in August this year.