Published in American Longwall Magazine

A leading practitioner of the approach is New South Wales-based consultancy Strata Control Technology (SCT), which has been working with computer modeling of longwall behavior for the past ten years. Much of the work has been in collaboration with the CSIRO (Commonwealth Scientific and Industrial Research Organization) and minesites.

Essentially computer simulation of the ground fracture process uses a base computer code to simulate the mechanics of failure and rock behavior. SCT uses the FLAC computer code developed by Itasca Consulting Group. As advanced users, SCT have applied detailed rock behavior characterization to match the actual ground response. This includes coupled fluid flow within the strata.

Site specific data is then entered into the model to generate a two-dimensional computer simulation along the central zone of the longwall panel.

To obtain a sound simulation of rock failure mechanics for a specific minesite, data such as rock strengths, stiffness, in-situ stresses, permeability and bedding plane characteristics have to be quantified.

With a long history of conducting and verifying computer simulations, SCT have found a strong correlation with actual field measurements. In other words – the simulations are very close to actuality.

“What has been found is that modeling capability is accurate and provides an understanding of what’s going on in the strata. We have therefore been able to adapt the approach to greenfields information and apply a predictive capability,” SCT managing director Winton Gale said.

Gale said he had conducted computer modeling of ground conditions for six new longwall operations in Australia.

The early and primary use of modeling was to identify and solve operational problems related to strata management. How to better use an existing longwall face, the effect of different roof support sets and yield loads on the strata, and the effects of longwall system health are all parameters that can be examined using the modeling.

The data can then provide an operator with an indication of what would happen should a certain course of action be taken.

For example, if a longwall has received inadequate maintenance over a period of time the effect on the surrounding strata of a decline in system health can be looked at.

Clearly an important factor regarding the modeling is the ability to validate the data and this is where SCT leads the field. The models developed by SCT have been validated at numerous minesites using monitoring of support pressure, subsidence and stress distributions to provide corroborating data for the models which have been applied to roadways, pillars and overburden caving.

Over the ten to fifteen years the company has worked with modeling, a consistent program of validation to confirm the mechanics has unfolded, representing a substantial achievement in the continuous development of a validated approach.

According to Gale several factors have enabled this - “good rock testing, good monitoring and an industry willing to back technology development, spend some research and development dollars and take a degree of risk”

An underlying theme in Gale’s work is the premise that rock failure initiates well ahead of the longwall face, an often neglected issue when considering face support design.

“The ground conditions are very different to what you’d expect if you are mining into fractured ground rather than mining into stable ground,” he said.

“For either situation different design parameters govern roof support design.”

SCT has studied a wide range of minesites, with conditions ranging from massive sandstone of 20-30m (66-98ft) thick to weak interbedded strata. Gale outlined examples of different results from two mines involving the influence of large scale caving on rock movements.

Modeling at the first mine was in a relatively weak environment with a weak to moderate strength roof section. The mined seam was three meters (9.8ft) thick within laminated siltstone and mudstone.

The computer simulation found the rock fracture was occurring well ahead of the mining face and was not related to the caving process behind the supports. The dominant rock failure modes recorded were shear fracture of rock and bedding accompanied by a relative absence of caving related failure within the strata.

“Overall, the computer simulation and the micro-seismic monitoring show very complementary results. The modeling provides an insight into the mode of failure represented by the micro-seismic activity. The overall results indicate that the rock fracture and the caving characteristics were pre-determined well ahead of the faceline,” Gale said.

A second minesite was characterized by cyclic caving in moderate strength massive strata. Cyclic caving is caving and/or support loading that occurs on a regular cyclic basis every 10-20m, accompanied by roof and rib fracture. Cyclic caving typically occurs within massive sandstone units with a massive nature (strong bedding plane characteristics) or which act as a massive unit within a much weaker rock mass.

Modeling of the process indicated that as mining moves under the massive unit it develops bending stresses.

At the point of collapse overburden rebound occurs where the overburden will displace on to the face and goaf area. The cyclic caving event is a combination of the formation of a fracture network and overburden convergence. The convergence of the overburden drives the convergence which may be experienced at the face.

This is a modification of the concept that gravity drop-out of an isolated block drives cyclic caving.

The severity of cyclic caving is related to the thickness of the massive unit, the material below it, the stiffness of the material in the overburden above the unit and face control procedures to maximize the stability of the immediate roof zone.

A third example of the application of computer modeling is the interaction of roof supports with ground movement around the face.

A review of computer models within weak ground indicated vertical stress and horizontal stress distributed above the support canopy played a major role in maintaining ground control around the longwall.

It has also been found the stress distribution above the canopy can vary on a shear by shear basis and is dependent on the integrity of the rock in the immediate roof above the canopy.

PART TWO: ... Click here to read on ...